A Dual Approach to Contract Remedies Michael D

YALE LAW & POLICY REVIEW

A Dual Approach to Contract Remedies Michael D. Knobler*

INTRODUCTION ...... 416

I. VALUE-BASED THEORY: THE CONTRACT AS OPTION...... 418 A. Economic Efficiency...... 419 B. Contract-Law Rules...... 420

II. PERFORMANCE THEORY: THE CONTRACT AS COMMITMENT ...... 424 A. How Performance Theorists React to Breach ...... 425 B. Societal Expectations ...... 427 C. The Deal to Which PartiesActually Consented, and Why That Matters ...... 433 D. The Search for a Middle Ground...... 441

III. How COURTS TACKLE THE QUESTION ...... 444 A. Bad-Faith Breach...... 444 B. Diminution in Value vs. Cost of Completion...... 447

IV. POTENTIAL SOLUTIONS: WHY BIFURCATING REMEDIAL APPROACHES MAKES SENSE ...... 448 A. Doing Nothing ...... 449 B. Notice ...... 449 C. Remedies Under a Performance Theory...... 450 D. CreatingDual Systems...... 453 i. How Contract Law Already Tailors Itself to Parties' Needs...... 453 2. Where To Apply the Option and Performance Theories...... 456 3. Should the System of Remedies Be Contractible?...... 457

CONCLUSION ...... 459

Yale Law School, J.D. expected 2012; Harvard University, A.B. 1985. I would like to thank Professors Rick Brooks, Greg Klass, and Daniel Markovits, Notes Editor Al- exandra Roth, and Daniel Hemel, Talia Kraemer, Robert Leider, Alex Platt, David Robinson, and Ellen Weis for their valuable comments.

415 YALE LAW & POLICY REVIEW 30:415 2012

INTRODUCTION

Numerous disagreements in contract law stem from the answer to a straightforward definitional question: What is a contract? Is it a mutual commitment to perform, or is it merely a promise to deliver the agreed-upon performance or pay damages, with each choice treated as equally acceptable? Is a contract about the right to obtain an agreed-upon good or service at a given time, or is it about the right to obtain an agreed-upon value at a given time?' Most people answer those questions differently than the law does. People overwhelmingly view breach as immoral, favor specific performance if breach occurs, and consider exclusively economic damages undercompensatory.' In contrast, courts typically treat breach as a rational option and deny nonbreaching parties specific performance and both nonpecuniary and punitive damages.3 The courts' approach seems targeted to the needs of businesses rather than individuals. 4 This disconnect between most people's beliefs and court enforcement undermines a foundational claim of contract law, that parties consent to a set of rules to guide their conduct and that courts enforce the parties' will.' Contract law zealously protects this consent principle in other areas, such as the statute of frauds, the doctrine of unconscionability, and the requirement for objective

1. Compare Louis KAPLOW & STEVEN SHAVELL, FAIRNESS VERSUS WELFARE 155-224 (2002) (arguing that the option to breach and pay damages maximizes social welfare), Oliver Wendell Holmes, Jr., The Path of the Law, io HARV. L. REV. 457, 462 (1897) ("The duty to keep a contract at common law means a prediction that you must pay damages if you do not keep it-and nothing else."), and Richard Pos- ner, Let Us Never Blame a Contract Breaker, 107 MICH. L. REV. 1349 (2009) (sup- porting the option theory), with Bailey v. Alabama, 219 U.S. 219, 246 (1911) (Holmes, J., dissenting) ("Breach of a legal contract without excuse is wrong conduct...."), Seana Shiffrin, Could Breach of Contract Be Immoral?, 107 MICH. L. REV. 1551 (2009) [hereinafter Shiffrin, Breach] (arguing for a performance-oriented approach), and Seana Valentine Shiffrin, The Divergence of Con- tract and Promise, 120 HARV. L. REV. 708 (2007) [hereinafter Shiffrin, Divergence] (same). 2. See infra Section II.B. 3. See infra Section I.B. 4. See infra Part I.

5. See Bidlack v. Wheelabrator Corp., 993 F.2d 603, 612 (7th Cir. 1993) (Cudahy, J., concurring) ("When the parties or the situation are such that those affected by the rule cannot bargain around it ... the rule may, in effect, supply terms to which the parties are bound but to which they have not explicitly agreed. In these circumstances, courts should try to select the default rule that best reflects the course the parties would have taken had they bargained freely.") (citing FRANK H. EASTERBROOK & DANIEL R. FISCHEL, THE ECONOMIC STRUCTURE OF CORPORATE LAW 15 (1991); RICHARD POSNER, EcONOMic ANALYSIS OF LAW 372 (3d ed. 1986)).

416 A DUAL APPROACH TO CONTRACT REMEDIES manifestations of intent.6 It seems incongruous, therefore, to frustrate consent when choosing remedies for breach. This Note does not choose sides in the decades-old value versus performance dispute. Instead, it suggests a both-and answer for a question to which others have given an either-or response. Rather than saying that all contracts are mutual commitments to perform, or that all contracts are merely promises to deliver the agreed-upon performance or pay damages, why not recognize that the idea of "contract" embraces both definitions? Some contracts fit in one category, some in the other. Just as the Uniform Commercial Code (UCC) sets forth different rules for contracts among merchants and for those among other parties,7 so contract law in general should-and perhaps does-recognize the differing expectations and understandings of different contractors. If business people typically contract for value and consumers typically contract for performance, the law should be tailored to reflect that difference. This conclusion follows whether one wishes contract law to maximize autonomy or efficiency. Not only does an exclusively option-oriented approach to contracts frustrate the will of those who wish to contract for performance, it also creates party error, which is a gap between the contract as understood by at least one of the parties and the contract as enforced by the courts.9 Party error leads to inefficient contracts.o Only by abandoning the one-size-fits-all approach can contract law maximize both efficiency and autonomy. I suggest two systems of remedies: a performance-oriented one for contracts to which at least one individual is a party and an option-oriented one for contracts involving only businesses. The performance-oriented approach would make more liberal use of specific performance and would make damages remedies more fully compensatory of both economic and nonpecuniary harms. The option-oriented approach would look much like current contract law, with its focus on compensating economic harms.

6. See infra notes 113-120 and accompanying text.

7. See, e.g., U.C.C. §§ 2-201(2), 2-205, 2-207(2), 2-209(2), 2-605(1)(b), 2-609(2) (1977). 8. See Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts:An Economic Theory of Default Rules, 99 YALE L.J. 87, 91 (1989) (defining a "tailored default" as a rule that attempts to provide parties "what they would have contracted for"); cf Hanoch Dagan, Pluralism and Perfectionism in Private Law 17 (Nov. 12, 2011) (unpublished manuscript) (on file with author) ("[D]espite the appeal of mon- ism's global coherence, value pluralism makes it reasonable and even desirable for law to adopt more than one set of principles and, therefore, more than one set of coherent doctrines."). 9. For a discussion of party error, see Ian Ayres, Regulating Opt-Out: An Economic Theory of Altering Rules, 121 YALE L.J. 2032 (2012). 10. See infra Section II.B.

417 YALE LAW & POLICY REVIEW 30:415 2012

This Note proceeds in four parts. Part I describes the efficiency justifications for the option theory and shows that many rules of current contract law reflect this approach. Part II describes the theory that contracts constitute moral commitments, examines empirical data revealing how well that theory matches ordinary people's understandings, and argues that a law of contracts that ignores those understandings violates the fundamental requirement that contracts be consensual. Part III analyzes case law to show how this central question-is contract about performance or value?-plays out in the doctrine of bad-faith breach and in the choice of remedies between cost of completion and diminution in value. Part IV analyzes potential responses to the problems created by the duality of contract theory and suggests bifurcating contract-law remedies along the performance versus value divide.

I. VALUE-BASED THEORY: THE CONTRACT AS OPTION

The prevailing economic view of contract law focuses on its role as the fa- cilitator of voluntary exchanges, presumed to be welfare-enhancing." Transac- tions that result in gains from trade should be encouraged, according to this theory, as parties' self-interest will make contractual exchanges Pareto efficient, leaving at least one party better off and neither worse off." Sometimes, though, the parties will discover after formation that performance would be more costly to the promisor than value-producing for the promisee. In such a situation, performance would lead to a net economic loss for society. The option theory presents a way of avoiding this outcome." The option theory treats contracts as if the parties had agreed to an additional clause giving the promisor an alternate way to perform. "[Olne may interpret the contract that reads 'Seller will build and deliver a cabinet to Buyer' to mean 'Seller will build and deliver a cabinet to Buyer, or else pay expectation damages, at Seller's option."" 4 Under this view, true breach never happens unless the promisor neither delivers nor pays damages, because paying damages acts as a mode of performance in a "dual performance hypothesis."'5 For clarity,

11. JOHN EDWARD MURRAY, JR., CONTRACTS 3 (1991). 12. Technically, this is true only for complete contracts, without bounded rationality or unforeseeability. See, e.g., Steven Shavell, Damage Measures for Breach of Con- tract, ii BELL J. EcON. 466, 467 (1980). For simplicity, I discuss contracts involving two parties, but the conclusions are generalizable to multiparty contracts. For a discussion of Pareto optimality, see JEFFRIE G. MURPHY & JULES L. COLEMAN, PHILOSOPHY OF LAW: AN INTRODUCTION TO JURISPRUDENCE 182-85 (rev. ed. 1990).

13. See KAPLOW & SHAVELL, supra note 1, at 172-213.

14. Id. at 192.

15. Daniel Markovits & Alan Schwartz, The Myth of Efficient Breach: New Defenses of the Expectation Interest,98 VA. L. REV. 1939 (2012).

418 A DUAL APPROACH TO CONTRACT REMEDIES this Note uses "performance" to mean complying with the stated terms of the contract, and it uses "breach" to mean not complying with those terms.

A. Economic Efficiency

The power of the option view of contract stems from its ability to deliver goods and services only when it is economically efficient to do so." The option to pay damages insures the promisor against the risk that some factor unfore- seen at formation will make the cost of performance exceed the value of that performance to the promisee. 7 Though it is easy to see the benefit to the promisor of being freed from a money-losing promise by paying damages totaling less than the amount she would have lost through performance, the benefit to the promisee comes into view only when one looks beyond the breach of contract to the terms of that contract and of other contracts for the same goods or services. Contracting parties set the terms so as to divide the gains from trade, de- fined as the difference between the value to the promisee and the cost to the promisor. The promisee will not pay more than her estimate of the value of performance, and the promisor will not charge less than the expected cost of performance. But the expected cost is based on the likelihood that the needed supplies will be available at certain prices. And the actual cost will be based not just on the prices of the necessary supplies, but also on the opportunity cost associated with the promisor forgoing other contracts she could have filled by rede- ploying the scarce resources-such as time or materials-required to perform her contract with the promisee. Imagine there is an 80% chance the total cost of performance will be $400 and a 20% chance it will be $500, depending on the cost of supplies and the value of forgone opportunities, so the expected cost of performance will be $420 = (o.8 x $400) + (0.2 x $5oo). If the promisee values performance at $450 and the gains from trade are divided evenly and performance is mandatory, the price will be $435. Ten such contracts, on average, would produce $300 in net gains, or $150 for each party. But when the promisor can choose between performance and paying expectation damages, performance will occur only when the cost of performance is lower than the value of that performance to the promisee." Ten

16. This assertion and the discussion that follows are based on Kaplow and Shavell's exposition. KAPLOW & SHAVELL, supra note i, at 172-213.

17. The cost of performance includes the opportunity cost of performance. For example, the cost of building and delivering a cabinet to the promisee includes not only things like material and labor but also the cost of any profit forgone by not selling the same cabinet to a buyer who was willing to pay more for it. 18. This claim assumes that expectation damages equal the value of the performance to the promisee. KAPLOW & SHAVELL, supra note 1, at 16o n.16. This often is not the case. See infra text accompanying notes 54-59.

419 YALE LAW & POLICY REVIEW 30 :415 2012 contracts, on average, would result in eight performances and two breaches, at a gross cost to the promisor of (8 x $400) + (2 x $450) = $4100 and a gross value to the promisee of to x $450 = $4500. If those gains remain evenly divided, the price drops to $430 per contract.'9 By giving the promisor the option to perform or pay damages, the option theory lowers the contract price and makes both the promisor and the promisee $5 per contract better off than they would have been had performance been mandatory.2 0 This example illustrates three big ideas. First, a performance that costs more to produce than it is worth to the promisee creates a net loss in social welfare, so both parties are better off with a contract theory under which such a performance does not occur. Second, if one includes opportunity costs in the costs of performance, then performance of a less-profitable contract creates a net loss in social welfare when breaching would have allowed the promisor to perform a sufficiently lucrative alternative. Third, rather than focus on remedies for breach from an ex post perspective, one must consider how remedies affect ex ante negotiations.' Proponents of an option theory argue that viewing performance as anything more than an option produces inferior economic results, and that any change in the rules directed at one party affects the costs that will be borne by the other.

B. Contract-LawRules

The system of rules governing contracts reflects this economic understanding and makes sense within the option framework. Strict liability and the rule against punitive damages, for instance, mean that the promisor generally incurs neither penalty nor forgiveness based on the reasons for breach. Treating damages as equivalent to performance removes any legal curiosity about the

19. (Gross value to the promisee - gross cost to the promisor) / number of contracts = ($4500 - $4100) / io = $40 in gains. Sharing those gains evenly at $20 for each party makes the price per contract equal the $410 cost plus the $20 promisor's share of the gains = $430. 20. Each party expects to gain $20 per contract under the option theory and $15 per contract if performance is mandatory.

21. See KAPLOW & SHAVELL, supra note i, at 203. 22. The argument presented here is simplified, and the conclusions, while main- stream, remain contested. For a broad critique of economic analysis of contract law, see Eric A. Posner, Economic Analysis of Contract Law After Three Decades: Success or Failure?,112 YALE L.J. 829 (2003). 23. See Globe Ref. Co. v. Landa Cotton Oil Co., 190 U.S. 540, 544 (1903) (Holmes, J.) ("[I]f a contract is broken the measure of damages generally is the same, whatever the cause of the breach."). The doctrines of impracticability and frustration of purpose are exceptions to this general proposition. RESTATEMENT (SECOND) OF CONTRACTS §§ 261, 265 (1981).

420 A DUAL APPROACH TO CONTRACT REMEDIES reasons the promisor chose one over the other. If breach is not wrong, the breacher need not justify it." A promisor's economic decision is presumptively rational, and allowing the promisor to choose between performance and so-called efficient breach allows for maximization of gains from trade. As Judge Richard Posner explained: The promisor may simply have discovered that his performance is worth more to someone else. If so, efficiency is promoted by allowing him to break his promise, provided he makes good the promisee's actual losses. If he is forced to pay more than that, an efficient breach may be deterred, and the law doesn't want to bring about such a result." Other familiar contract rules-about specific performance, foreseeability, and mitigation-also reflect an option view." Rules governing court-ordered remedies present the most straightforward evidence that the existing law of contracts favors the option theory. Courts grant specific performance only when damages would be inadequate, 7 and the rarity of specific performance outside the realm of real-estate contracts attests to the perceived adequacy of damages.2 The option theory holds such sway over the choice of remedy that even if a contract calls for specific performance, courts may refuse to enforce it and may impose damages instead. 9 Policy reasons for this include the administrative issues surrounding the monitoring and enforcement of court orders, the inalienability

24. See Posner, supra note 1, at 1350-51 (absolving breachers of blame and connecting the option theory to the strict-liability rule). 25. Patton v. Mid-Continent Sys., Inc., 841 F.2d 742, 750 (7th Cir. 1988).

26. Shiffrin, Divergence, supra note 1, at 722-27. 27. See, e.g., Van Wagner Adver. Corp. v. S & M Enters., 492 N.E.2d 756 (N.Y. 1986); RESTATEMENT (SECOND) OF CONTRACTS § 359. Though "modern contract law generally has liberalized the availability of specific performance," Melvin Aron Ei- senberg, The Emergence of Dynamic Contract Law, 88 CALIF. L. REV. 1743, 1761 (2000), there is evidence that courts embraced liberalization more slowly than commentators would have preferred, see U.C.C. § 2-716 cmt. 1 (2004) (stating that the Uniform Commercial Code (UCC) "seeks to further a more liberal attitude than some courts have shown in connection with the specific performance of contracts of sale"); see also Weathersby v. Gore, 556 F.2d 1247, 1258 (7th Cir. 1977) (quoting the UCC comment but holding that the Mississippi Supreme Court would make specific performance available more restrictively). 28. See, e.g., CIMA Capital Partners, LLC vs. PH Cellular, Inc., 69 So. 3d 293, 294 (Fla. Dist. Ct. App. 2010) (noting the rarity of specific performance and reversing an order for it). 29. See Avery Katz, Virtue Ethics and Efficient Breach, 45 SUFFOLK U. L. REV. 777, 781 & n.14 (2012) (citing Stokes v. Moore, 77 So. 2d 331 (Ala. 1955) (refusing to enforce the agreement's terms but noting that they served as evidence of the inadequacy of money damages)).

421 YALE LAW & POLICY REVIEW 30:415 2012 principles that many specific performance orders would contravene, and the worry that the threat of enforcing specific performance will lead to the promisee obtaining supercompensatory damages.30 Also, many contracts have a temporal element that renders post-breach specific performance either impossible or im- practical.3 No matter how brilliant judges may be, they possess neither the time nor the tools to administer specific performance of every breached contract that comes before them. No matter how much Americans want to encourage promisors to meet their contractual obligations, court-mandated indentured servitude offers a cure far worse than the disease.32 No matter how badly an airline ticketholder wants to ride on an overbooked flight, the plane has only so many seats. The pragmatic problems with specific performance are visible to plaintiffs, too: A 1990 article said that in Europe, where specific performance was easy to get, it was seldom sought." Not only do courts usually reject specific performance, but they also reject assessing damages in ways that favor performance over breach. Consider the Hadley rule, which limits promisors' liability for breach to the consequential damages foreseeable at formation. 4 Commentators commonly describe the rule as information-forcing; promisees can protect their rights to full recovery only by disclosing relevant information about their needs and expectations, including the extent to which they are relying on performance. This explanation makes perfect sense under the option theory; if breach is an acceptable option, then the promisor is entitled to know at the time of formation all existing relevant information about potential damages. But if people contract for performance, perhaps damages for breach should be based on what was foreseeable at

30. See Randy E. Barnett, Contract Remedies and Inalienable Rights, 4 Soc. PHIL. & POL'Y 179, 181 (1986). But see Alan Schwartz, The Case for Specific Performance, 89 YALE L.J. 271, 292-98 (1979) (arguing against the conventional wisdom).

31. See KAPLOW & SHAVELL, supra note I, at 163 n.21 (citing the example of the overbooked airliner).

32. See RESTATEMENT (SECOND) CONTRACTS § 367(1). But see Nathan B. Oman, Spe- cific Performance and the Thirteenth Amendment, 93 MINN. L. REV. 2020 (2009) (arguing that specific performance would rarely amount to indentured servitude and that application of specific performance in the employment context would not directly conflict with the Thirteenth Amendment). 33. Alan Schwartz, The Myth that Promisees Prefer Supracompensatory Remedies: An Analysis of Contractingfor Damage Measures, ioo YALE L.J. 369, 389 (1990).

34. Hadley v. Baxendale, (1854) 156 Eng. Rep. 145 (Exch.); see Ayres & Gertner, supra note 8, at 1o1-o4; Nathan B. Oman, The Failure of Economic Interpretations of the Law of Contract Damages, 64 WASH. & LEE L. REV. 829, 843-44 (2007). But see Posner, supra note 22, at 836-37 (explaining that the absence of the Hadley rule also would be information-forcing, as promisees with lower-than-average reliance interests would disclose those interests and thereby obtain lower prices thanks to the lower-than-average damages they would collect in the event of breach).

422 A DUAL APPROACH TO CONTRACT REMEDIES breach, under the theory that calculating damages in that way would provide the most accurate measurement of the harm incurred." But changing the foreseeability rule might reduce efficiency. Though calculating damages based on what is foreseeable at breach produces more equitable results and stronger incentives for performance ex post, it produces higher prices ex ante. Even with a performance-oriented contract-law system, breach will sometimes occur. When damages are calculated based on what is foreseeable at breach, the promisee has an incentive to overinvest in reliance on performance because she need not factor in the risk of breach. 6 Such overinvestment is cost-free to the promisee ex post but costly to promisees in general ex ante because promisors, aware that they will be liable for all of that inefficient investment, will adjust their price demands accordingly. This argument is analogous to the efficiency discussion in Section II.A: A performance-oriented rule leads to higher cost. Of course, it also leads to a higher rate of performance. A performance-based theory of contract would compel the promisor to perform or, failing that, to do the best she can.3 7 But contract law requires the promisee, not the promisor, to mitigate damages.3' This requirement is consistent with the option-theory notion that breach is equally as acceptable as performance. If neither side deserves blame for the lack of performance, why not assign mitigation to the promisee, who best understands her own needs and thus arguably holds a comparative advantage for obtaining the appropriate sub- stitute? In contrast, under a performance theory, the burden would follow the blame.

35. See Shiffrin, Divergence, supra note 1, at 724. What is foreseeable at breach is not always observable at breach. For example, breach can lead to damages for unreal- ized future profits. Though those profits must be reasonably certain, that certainty need not exist at breach but only at the time damages are assessed, and even then the profits themselves are not observable. United States v. Penn Foundry &Mfg. Co., 337 U.S. 198, 2o8 (1949) (requiring only "a sufficient basis for estimating [un- realized profits] with reasonable certainty"). 36. For a discussion of the problem of overinvestment in reliance, see Melvin A. Ei- senberg & Brett H. McDonnell, Expectation Damages and the Theory of Overreli- ance, 54 HASTINGS L.J. 1335 (2003). Eisenberg and McDonnell argue that the failure of expectation damages to compensate fully for breach might be a feature, not a flaw, because the undercompensation makes overinvestment in reliance a costly strategy. Id. at 1357-62; cf George Wyeth, Regulatory Compensation and the Tak- ings Clause, 91 Nw. U. L. REV. 87, 101-02 (1996) (discussing the concepts of foreseeability and overinvestment as they apply to compensation under the Takings Clause). 37. See Shiffrin, Divergence, supra note 1, at 725. 38. RESTATEMENT (SECOND) OF CONTRACTS § 350 & illus. 5-8 (1981) (requiring "appropriate efforts" from the promisee).

423 YALE LAW & POLICY REVIEW 30 :415 2012

Courts do sometimes act as if they are applying the performance theory. 9 For example, when breach of a contract accompanies breach of a fiduciary duty, the breaching party may be forced to disgorge any ill-gotten gains.40 Colorado common law distinguishes between "mere breach of a contract" and "wrongdo- ing [that] is intentional or substantial," which trial courts have the discretion to remedy by ordering disgorgement. 41 Those situations remain exceptions, however. If the law treated breach as a wrong and performance as an obligation, many rules governing remedies would require revision. The choice of rules generally reflects the law's acceptance of the option theory.

II. PERFORMANCE THEORY: THE CONTRACT AS COMMITMENT

Performance theorists see contracts as promises and therefore argue that breach is wrong because breaking promises is wrong. 42 Some see a contract as granting the promisee a property right that the promisor cannot justifiably re- voke without permission. 43 Performance theorists believe the state's interest in contract enforcement extends beyond mere wealth maximization. The state should avoid "harms to the practice of undertaking voluntary obligations and harms resulting from its abuse." 44 Performance theorists would add two words to flip the 1992 Bill Clinton slogan; for them it's not just the economy, stupid. It's about right and wrong. Their argument benefits from examples of people sacrificing money in pursuit of other values. 45

39. Steven W. Feldman, Betrayal and Exploitation in Contract Law: A Comment on Breach is for Suckers, 63 VAND. L. REV. EN BANC 171, 173-74, 181-82 (2010); see infra Part III and text accompanying notes 78-81.

40. See, e.g., Snepp v. United States, 444 U.S. 507, 515-16 (1980) ("conforming relief to the dimensions of the wrong" by awarding the CIA the profits from a book written by a former agent who breached a contractual requirement that he clear the book's contents with the CIA before publication). 41. EarthInfo, Inc. v. Hydrosphere Res. Consultants, 9oo P.2d 113, 119 (Colo. 1995) (en banc). For more on disgorgement, including examples of courts in various states applying the remedy, see Melvin A. Eisenberg, The Disgorgement Interest in Con- tract Law, 105 MICH. L. REV. 559 (2006).

42. CHARLES FRIED, CONTRACT AS PROMISE 17 (1981). 43. See, e.g., Randy Barnett, A Consent Theory of Contract, 86 COLUM. L. REV. 269, 297-98 (1986); Daniel Friedmann, The PerformanceInterest in Contract Damages, InI L.Q.R. 628, 637-38 (1995). 44. Joseph Raz, Promises in Morality and Law, 95 HARV. L. REV. 916, 938 (1982). 45. See, e.g., Yochai Benkler, Law, Policy, and Cooperation, in GOVERNMENT AND MARKETS: TOWARD A NEW THEORY OF REGULATION 299, 300 (Edward J. Balleisen & David A. Moss eds., 2010) (stating that less than one-third of humans act as wealth maximizers); Ernst Fehr & Simon G5chter, Altruistic Punishment in Hu-

424 A DUAL APPROACH TO CONTRACT REMEDIES

The remainder of this Part explores the implications that people's non-economic motivations hold for contract law. Section A discusses one of the most difficult challenges facing the performance theorist: If contracts oblige promisors to perform, how does this premise inform the choice of remedy for breach? Section B presents experimental and real-world evidence that people conceive of contracts as creating obligations instead of options. Section C eval- uates the implications of that empirical evidence for the legitimacy of contract law. Section D looks at an attempt to create a theoretical middle ground between the option and performance approaches.

A. How Performance Theorists React to Breach

Seana Shiffrin tells the story of waiting in vain for a plumber who fails to arrive for a scheduled repair job.46 The breach results in frustration, annoyance, and anger-things for which contract law will not compensate 47 -but Shiffrin shows that the problem is the breach itself rather than how and whether it is compensated.48 She asks how a reader would feel if the plumber, finally arriving on a later day, presented a check or a discount as full compensation for the time wasted waiting for the plumber's arrival.49 Resentment would persist, not from "1emotional inertia" but from an actual harm that money does not adequately compensate."o If, as this story suggests, performance and damages sometimes are incommensurable, what would a performance-based theorist do in the event of breach? The answers differ. Specific performance makes sense if one views a contract as obliging performance, and some performance-based theorists embrace the idea, with limitations for pragmatic and inalienability reasons.5 ' Randy Bar- nett would reverse the specific-performance presumption and burden of proof; instead of requiring a promisee to prove that damages are inadequate, Barnett would require the breacher to prove that damages are adequate." Charles Fried and Daniel Friedmann favor expectation damages on the theoretical ground

mans, 415 NATURE 137, 137 (2002) (showing that people will spend money to punish others who have contravened societal norms, even if such expenditures will not benefit the spenders). 46. Shiffrin, Breach, supra note 1,at 1564. 47. See Schwartz, supra note 30, at 276. 48. Shiffrin, Breach, supra note i, at 1564. 49. Id. 50. Id. 51. See supra text accompanying notes 30-32.

52. Barnett, supra note 30, at 183.

425 YALE LAW& POLICY REVIEW 30:415 2012 that such damages place the promisee in as good a position as she would have been had the contract been performed." Reality, however, does not live up to that theory. Expectation damages can be undercompensatory for several reasons. Damages must be foreseeable,5 4 as- certainable," and unavoidable."6 Recovery for emotional disturbance is available only in limited circumstances in which the nature of the contract or the breach made such a disturbance likely.17 The promisee loses the difference between the subjective and objective value of performance.'" Add in the effects of the Ameri- can system, with each party bearing its own litigation costs, and the result is that "the nonbreaching party is rarely fully compensated." 9 This undercompensation is not entirely offset by a reduction in the price charged to promisees. Because unsophisticated parties underestimate enforcement error, or fail to recognize its existence, they do not get the reduction in price that they should."o But even if they did, there would still be two distribu- tive justice problems. First, the benefits from the ex ante price reduction would go to a different group than the costs from the ex post undercompensation for breach. Where undercompensation exists, promisees whose contracts are

53. KAPLOW & SHAVELL, supra note 1, at 161 n.8. 54. See Hadley v. Baxendale, (1854) 156 Eng. Rep. 145 (Exch.); RESTATEMENT (SE- COND) OF CONTRACTS § 351 (1981).

55. RESTATEMENT (SECOND) OF CONTRACTS § 352. 56. Id. § 350. 57. Id. § 353; 24 SAMUEL WILLISTON, A TREATISE ON THE LAW OF CONTRACTS § 64:7 (Richard A. Lord ed., 4th ed. 2002). 58. See Marco J. Jimenez, The Value of a Promise: A UtilitarianApproach to Contract Law Remedies, 56 UCLA L. REV. 59, 97 (20o8); Thomas S. Ulen, The Efficiency of Specific Performance: Toward a Unified Theory of Contract Remedies, 83 MICH. L. REV. 341, 375-76 (1984). 59. Scott Baker & Kimberly D. Krawiec, Incomplete Contracts in a Complete Contract World, 33 FLA. ST. U. L. REV. 725, 737 (20o6) (citing Robert E. Scott & George G. Triantis, Embedded Options and the Case Against Compensation in Contract Law, 104 COLUM. L. REV. 1428, 1448-49 (2004)); see also Eisenberg & McDonnell, supra note 36, at 1358 (estimating that "at the time a risk-neutral promisee invests in beneficial reliance, the expected value of damages based on beneficial reliance is unlikely to often exceed seventy to eighty percent of perfect-world damages, even without regard to litigation costs"). But see Ian Ayres & Kristin Madison, Threat- ening Inefficient Performance of Injunctions and Contracts, 148 U. PA. L. REV. 45 (1999) (arguing that court-ordered damages sometimes overcompensate promisees and that promisors will threaten to deliver inefficient performance as a means of bargaining for a lower damages settlement). 60. See Barnett, supra note 30, at 183.

426 A DUAL APPROACH TO CONTRACT REMEDIES breached subsidize promisees whose contracts are performed.' Second, the burden of that subsidy falls unevenly even among the victims of breach, because the extent of undercompensation varies based on the circumstances of the breach. Consider two couples who have contracted with a hotel to host their wedding receptions. The uncompensated emotional distress harm from breach is likely to be much higher for the couple whose contract is breached a week before the wedding than for the couple whose contract is breached months earlier. Both received the same ex ante price reduction from the undercompensation of breach, and couples whose contracts were not breached received that price reduction too. A solution lies in granting specific performance where possible or in granting damage awards that compensate non-economic harms as well as economic ones. If the insight of performance theorists is that the promisee is not indifferent between performance and expectation damages, the answer lies in assuring either performance or a level of damages that most closely reflects the harm from breach." Anything less thwarts the expectations of those who have contracted for performance and thereby calls into question whether they are being governed by contractual rules to which they have consented.

B. Societal Expectations

"Persons with common sense-that is, those who have not taken a first-year contracts class (or been counseled by a lawyer who has)-would natu- rally assume" that the performance and not the option theory of contract holds true, wrote Barnett, who has not only taken a first-year contracts class but has taught many as well.3 Even a noted advocate of efficient breach conceded that people's intuitions about contracts match the performance-based theory. "My experience, and I suspect the reader's, has been that most individuals react to breach . . . as having an ethically incorrect aspect," Steven Shavell wrote.6 4 He confirmed this notion with a survey that asked people's assessment of a hypothetical kitchen renovator breaching a contract with a homeowner because an unexpected rise in costs

61. Undercompensation of breach lowers costs for promisors, who, in a competitive market, share that reduction with all promisees via the ex ante price of a contract. For other quantitative implications of breach, see infra notes 135-141 and accompanying text. 62. See infra Section IV.C for proposed remedies under a performance-based theory of contract. 63. Barnett, supra note 30, at 183. Barnett is a professor at Georgetown University Law Center. 64. Steven Shavell, Why Breach of Contract May Not Be Immoral Given the Incom- pleteness of Contracts, 107 MICH. L. REV. 1569, 1579 (2009).

427 YALE LAW & POLICY REVIEW 30:415 2012 would cause the renovator to lose money on the job." Shavell offered a scale for responses, with "definitely unethical" as i, "somewhat unethical" as 2, "neither ethical nor unethical" as 3, "somewhat ethical" as 4, and "definitely ethical" as 5-66 The 2.41 mean result shows the respondents disfavored the proposed economically efficient breach.6 7 Efficient breach and the option theory of contract fared no better in surveys conducted by Tess Wilkinson-Ryan and Jonathan Baron. 68 In one of three scenarios, a floor refinisher breached a contract so that he could accept a high- 6 er-paying job. 9 On a i to 7 scale with 4 being "somewhat immoral" and 7 being "extremely immoral," participants' average ranking of breach was higher than 5.70 More than three-quarters of the members of an Internet panel said that the promisor should perform rather than breach and pay damages, though it was clear in each scenario that the promisor would be economically better off by breaching.7' More than two-thirds of the panelists said a court should order specific performance. 2 Considerations that are economically equivalent-opportunity cost versus other costs of performance-were not morally equivalent to the panel. When asked to set damages for hypothetical instances of breach, people reacted more punitively to breaches motivated by the opportunity to gain money than to breaches motivated by the opportunity to avoid losing money.73 This effect occurred even when the respondents were not making side-by-side comparisons; those who saw only the "opportunity to gain money" scenario assessed higher damages than those who saw only the "opportunity to avoid losing money" scenario. 74 That outcome is inconsistent with an option theory of contract, in which there is no obligation for the promisor to perform and hence no need for the adjudicator to investigate the reasons for nonper-

65. Steven Shavell, Is Breach of Contract Immoral?, 56 EMORY L.J. 439, 453-55 (20o6). 66. Id. at 453. This scale seems strange, as it combines the answers to two questions: Does this action have an ethical dimension, and, if so, is the action ethical? Is a "no" answer to the first of those questions (a 3 on Shavell's scale) worse than a "yes" answer combined with the conclusion that the renovator's actions were only "somewhat ethical" (a 4)? That is, can an action be somewhat ethical without also being somewhat unethical? 67. Id. 68. Tess Wilkinson-Ryan & Jonathan Baron, Moral Judgment and Moral Heuristics in Breach of Contract,6 J. EMPIRICAL LEGAL STUD. 405 (2009). 69. Id. at 418.

70. Id. at 419. 71. Id. at 420.

72. Id. 73. Id. at 414.

74. Id.

428 A DUAL APPROACH TO CONTRACT REMEDIES

formance. But under the performance theory, where paying economic damages does not absolve the promisor of responsibility for failing to fulfill an obligation, the reason for breach becomes relevant to the determination of damages. That relevance becomes clear in light of another Wilkinson-Ryan and Bar- on finding. Their survey participants awarded higher-than-expectation damages to promisees, and the multipliers depended on the circumstances of the breach.75 This outcome suggests that some of the damages were intended to be punitive, though theory and practice supposedly limit punitive damages to the realm of tort. Wilkinson-Ryan and Baron showed how counterintuitive that doctrinal limitation is by presenting their participants with tort scenarios that paralleled the contract ones. In each scenario, a third party's negligence resulted in the same damage to the promisee as a contract breach would have inflicted. In each tort scenario, participants assessed lower damages than in a comparable contract scenario.76 Wilkinson-Ryan and David Hoffman explain this effect empirically and theoretically as a recognition of the harm that comes from being exploited and betrayed."7 Real-world examples back up these experimental results. Given the opportunity to punish contract breachers, juries do?7 Marc Galanter studied court cases in the nation's seventy-five largest counties in 1992 and found that only four percent of winning plaintiffs in tort jury trials were awarded punitive damages, a third of the rate for winning plaintiffs in contract jury trials.79 Because a higher percentage of contract plaintiffs than tort plaintiffs won their cases, contract plaintiffs (successful and unsuccessful combined) got punitive damage awards at four times the rate of tort plaintiffs.o Galanter provided no analysis of the percentage of cases in which punitive damages were legally available; many states limit punitive damages in contracts matters to cases involving an accompanying tort. But his numbers do show that juries were far more likely to award punitive damages in what he called "uphill" contract cases-involving employment, fraud claims, and suits by buyers-than in "downhill" cases- such as nonpayment of rent, default on payment owed a seller, and mortgage foreclosure."'

75. Id. at 420 tbl.i. 76. Two of the three negligence awards averaged less than the contractual expectation damages. Id. 77. Tess Wilkinson-Ryan & David A. Hoffman, Breach Is for Suckers, 63 VAND. L. REV. 1003 (2010). 78. Marc Galanter, Contract in Court: Or Almost Everything You May or May Not Want To Know About Contract Litigation, 2001 Wis. L. REV. 577, 604-05. 79. Id. 80. Id. 81. Id. at 593, 604. Though this finding might suggest that people believe in the performance theory of contract only when the promisor is a faceless corporation and

429 YALE LAW& POLICY REVIEW 30:415 2012

People overwhelmingly reject the option approach to mortgage contracts. Surveys taken in 20o8 and 2009 showed that eighty-one percent of Americans believed that homeowners with the means to pay their mortgages had a moral duty to continue paying." Brent T. White argued that American politicians, media, and culture reinforce that belief and "cultivate fear, shame, and guilt" in homeowners contemplating foreclosure." Those emotions lead homeowners to believe that they bear moral as well as legal responsibility for their mortgages and must continue paying them even when rational self-interest would dictate otherwise.84 To White, the view of mortgage contracts as promises to perform acts as a tool of the mortgage-loan industry to shift the risk of the loan onto homeowners." In contrast, lenders treat their own mortgages as nothing more than option contracts and simply walk away from their properties when such an action makes financial sense." Though the comparison is imperfect because commercial real-estate loans tend to be nonrecourse and home loans are overwhelmingly recourse,7 this might in practice be a distinction without a differ-

the promisee is an individual, evidence shows otherwise. For example, the option theory of contract would predict that unenforceable clauses of contracts would not affect promisors' behavior, as they would know that breach would have no legal consequences. But people continued writing racially restrictive covenants after they became unenforceable because "it may have a moral or psychological effect upon a prospective purchaser encompassed by the restrictive covenant." CLEM- ENT E. VOSE, CAUCASIANS ONLY 228 (1959) (quoting Kenneth D. McCasland, Comment, PracticalEffects in Tennessee of the Non-Enforceability of Restrictive Ra- cial Covenants, 2o TENN. L. REV. 679, 681 (1949)). In other words, a nontrivial number of people who were not legally bound felt some sense of obligation to carry out the promises in their contracts. 82. Luigi Guiso, Paola Sapienza & Luigi Zingales, Moral and Social Constraints to Strategic Default on Mortgages 19 (Nat'l Bureau of Econ. Research, Working Pa- per No. 15145, 2009), available at http://www.nber.org/papers/wl5145.pdRnew window=1. 83. Brent T. White, Underwater and Not Walking Away: Shame, Fear, and the Social Managementof the Housing Crisis, 45 WAKE FOREST L. REV. 971, 997 (2010). 84. CBS's 60 Minutes interviewed an unemployed woman who said she was cutting back on blood pressure medications and food so she could continue to make mortgage payments despite owing sioo,ooo on a house worth $50,ooo. 6o Minutes (CBS television broadcast Dec. 18, 2011). Another woman said she continued to make payments on a mortgage higher than the worth of her house because "Itihat's just how I was raised. Once you, you know, you sign it, ... it's a contract[;] you uphold what you can for as long as you can." Id. 85. White, supra note 83, at 997.

86. Id. at 1009-12, 1023. 87. Curtis Bridgeman, The Morality of Jingle Mail: Moral Myths About Strategic De- fault, 46 WAKE FOREST L. REV. 123, 127 (2011).

430 A DUAL APPROACH TO CONTRACT REMEDIES

ence because-for practical and legal reasons-mortgage lenders rarely seek de- ficiency judgments."8 Under the Holmesian notion of the option theory of contract, the parties should care about an accurate prediction of what happens in the event of breach rather than a literal interpretation of the contract's text.89 Yet the surveys discussed here show otherwise. Curtis Bridgeman solves this apparent conflict by attacking the option theory.90 To Bridgeman, people act rationally when they respond to their moral intuitions about the responsibility to pay their mortgage debts rather than default and accept the legal consequences. Those favoring an option theory of contract, he says, reason fallaciously when they equate a legal action with a moral one: What you can do is often different than what you should do, and being willing to accept an action's consequences does not make a wrong action right.9 ' A Holmesian "bad man"' or a law and economics proponent might see no difference between a fine and a price, but for most people the disapproba- tion accompanying the former distinguishes it from the latter.93 Being willing to serve the prison term for murder does not confer the right to murder, and being willing to pay the remedy for breach does not confer the right to breach. "[IIt is a mistake," Shiffrin writes, "to regard the available remedy for breach of contract as fully exhausting the content of the duty that one assumes when one promises (or when one contracts)." 94 But there are reasons to suggest that Bridgeman's analogy between murder and contract law does not hold. For example, some people choose to risk receiving a parking ticket rather than put money in the meter. Those people are calculating that in the long run the price of the tickets they have to pay, multiplied by the number of tickets, will be lower than the price of paying every time they park to avoid having the meter expire before they move their cars. Perhaps they also factor in the inconvenience of carrying enough change for the meter and

88. See White, supra note 83, at 985 & n.53. 89. See Holmes, supra note i, at 457, 459, 462, 465 (defining the object of the study of law to be acquisition of the ability to make accurate predictions about legal consequences). 9o. Bridgeman, supra note 87, at 145-48.

91. See also Daniel Friedmann, The Efficient Breach Fallacy, 18 J. LEGAL STUD. 1, 15-16 (1989) (arguing for a distinction between lawful and permissible appropriation without the owner's consent, as in Vincent v. Lake Erie Transp., 124 N.W. 221 (Minn. 1910), and wrongful conduct).

92. Holmes, supra note 1, at 459 ("If you want to know the law and nothing else, you must look at it as a bad man, who cares only for the material consequences which such knowledge enables him to predict .... ).

93. See Richard R.W. Brooks, The Efficient Performance Hypothesis, u6 YALE L.J. 568 (2oo6). 94. See Shiffrin, Breach, supra note 1, at 1554.

431 YALE LAW& POLICY REVIEW 30:415 2012

even the odds that the city will enforce payment of a ticket. This is a purely economic calculation in the eyes of most people who make it, not a moral one. With some exceptions, such as handicapped spaces and the blocking of fire hy- drants, parking ordinances carry limited moral content. Contracts, one can argue, should be viewed similarly. The parties, in this view, have created morally neutral private law that they are free to violate when and if it becomes economically rational to do so. The notion that choosing to pay damages rather than to carry out performance is analogous to choosing to accept a prison sentence rather than to forgo commit- ting murder presupposes the very thing Bridgeman's analogy purports to prove: Parties to contracts face moral duties to deliver performance. A more complete analogy might suggest that private law mirrors public law. Just as some breaches of the social contract-such as not feeding the meter-constitute amoral rather than immoral choices, some private contracts carry with them no moral obligation to deliver performance. Tax lawyers know the difficulty of setting public policy that draws a line between laws that impose moral obligations and laws that impose strictly economic penalties. In Tank Truck Rentals, Inc. v. Commissioner,5 fines that truck companies paid for violating Pennsylvania's weight limit laws were not deductible as ordinary and necessary business expenses, though the Court admitted that the companies found it "economically impossible" to obey Pennsylvania's weight limit, which was 15,000 pounds lower than the limits in surrounding states.9' The company argued that the fines were analogous to revenue tolls, but the Court ruled that they were punitive.97 In contrast, Judge Learned Hand ruled in Jerry Rossman Corp. v. Commissioner that payments of overcharges from violations of a price control law were deductible, in part because they were restitutional rather than punitive.9' Applied to contract law, where there generally are no punitive damages, this distinction might suggest that breach should be treated as an acceptable business decision rather than as something the law should discourage. But Jerry Rossman can stand for a contrary proposition, too. Judge Hand considered whether deductibility might undermine deterrence and decided that for an inadvertent violation it would not.99 Just as Judge Hand wrote that "there are 'penalties' and 'penalties,' and ... some are deductible and some are not,""oo perhaps there are "contracts" and "contracts," and some create moral obligations and some do not.

95. 356 U.S. 30 (1958). 96. Id. at 32-33. 97. Id. at 36.

98. 175 F.2d 711, 712 (2d Cir. 1949). 99. Id. at 713-14. 1oo. Id. at 713.

432 A DUAL APPROACH TO CONTRACT REMEDIES

If so, the question becomes how to distinguish the two categories. One answer lies in the identities of the parties. For Bridgeman, the moral duty to pay one's mortgage vanishes when the "one" in question is a "nonhuman legal en- tit[y]."'o' As a result, although the option theory of contract may be dead to Bridgeman when it comes to human parties, it appears to be alive and well for limited liability companies. But his justification-that anyone agreeing to a contract with a limited liability company knows the risk involved-is circular. If everyone's knowledge of the background norm can serve as justification for the choice of that background norm, then with enough notice any choice of background norm will do.'0o Bridgeman might contend that the background norm is not a matter of choice but rather a matter of cultural expectations; he does not question the existence of one background norm for humans-parties are under an obligation to perform-and a different background norm for nonhuman legal entities-parties may act in their self-interest. By identifying and accepting those dual norms, Bridgeman in essence argues for a dual theory of contract. Articulating such a theory is the project of this Note.

C. The Deal to Which PartiesActually Consented, and Why That Matters

Simply showing that the performance-based theory aligns with most people's expectations does not, in itself, prove that contract law should be performance-oriented.o 3 People might not be entitled to receive what they expect.0 4 If the performance theory merely springs from unsophisticated parties' notions of morality, Holmes would object that conflating law and morality represents what we now call a category error: A contract is a legal document; morality is not a legal concept.o' To the extent that law and morality use the same terms,

loi. See Bridgeman, supra note 87, at 149. 102. Cf KAPLOW & SHAVELL, supra note 1, at 163-64 (arguing that those who breach contracts are not breaking promises if the background rule is "perform, or pay expectation damages"); Richard Craswell, Contract Law, Default Rules, and the Philosophy of Promising, 88 MICH. L. REV. 489, 490 (1989) (asserting that "the fi- delity principle [of contracts] is consistent with any set of background rules because those rules merely fill out the details of what it is a person has to remain faithful to"). 103. Nor, for that matter, does showing that the option theory is economically efficient prove that businesspeople prefer it. "We cannot move so quickly from the fact that rationally self-interested parties should prefer dual-performance commitments to the conclusion that sophisticated parties do prefer them." Gregory Klass, To Perform or Pay Damages, 98 VA. L. REV. 143, 146-47 (2012) (responding to Mar- kovits &Schwartz, supra note 15). In the absence of empirical evidence to the contrary, however, this Note makes the move that Klass questions. 104. Note that, despite their name, "expectation damages" often do not amount to what people actually expect. 105. See Holmes, supra note 1, at 462.

433 YALE LAW & POLICY REVIEW 30:415 2012 one must remember that those terms mean different things in these different contexts.o' Shiffrin's comment that contract law and ordinary promises use identical language thus carries no weight with a Holmesian.o 7 Even Shiffrin does not call for contract law to be coextensive with notions of promissory morality, just compatible with them.' There is and must be a gap between what is legal and what is moral in contract law.o 9 It does not follow, however, that this gap has no consequences for contract law's legitimacy. In fact, doctrine that makes the gap wider than necessary undermines the economic and moral foun- dations of contract law. First, economic justifications for contract law begin with the notion that contracts improve social welfare because they are Pareto efficient.1 o But if contract law frustrates people's expectations-if promisees are not, in fact, indifferent to obtaining damages rather than performance-then people will enter into contracts that make them probabiistically worse off at the instant of formation. Some contracts will even be Kaldor-Hicks inefficient" with respect to the contracting parties, depending on the likelihood of breach, the promisee's valuations of damages and performance, the price of the contract, and the promisor's expected return. This outcome occurs as a result of promisees' misunderstand- ing about how the law shifts the relevant demand curve. For example, recall the contract discussed in Section I.A, with the promisor paying $430 per unit and getting $450 in value when the promisor performs, with a twenty-percent chance of breach. Imagine that damages minus uncom- pensable harms and attorney's fees will be only $2 per contract, and that the promisee could have gotten a return of $19 per $430 by investing the money rather than entering a contract. Ten contracts, on average, return a value of $3604

106. Id. at 463. 107. See Shiffrin, Divergence, supra note 1, at 721. 108. Id. at 713-17. 109. For example, you may morally bind yourself to perform an act that the government cannot compel you to perform, whether for First Amendment or Thirteenth Amendment reasons. Also, contract law is limited to what is administrable, while morality is not. no. See supra text accompanying notes 11-14. in. Kaldor-Hicks efficiency is a measure of wealth maximization that ignores the way in which the wealth is distributed. A transaction is Kaldor-Hicks efficient "when- ever the winners win more than the losers lose, in the sense that, if the winners compensated the losers to their satisfaction, the winners would still be better off than they were before the change." Guido Calabresi, The Pointlessness of Pareto: CarryingCoase Further,100 YALE L.J. 1211, 1221 (1991). Note that even where there is no party error, negative externalities can make some contracts Kaldor-Hicks inefficient with respect to the society as a whole. My claim is that party error leads to more wealth-destroying contracts than would otherwise exist.

434 A DUAL APPROACH TO CONTRACT REMEDIES

= (8 x $450) + (2 x $2) at a cost of $3440 = 8 x $430, for a net return of $164. The promisee was expecting performance, with a value of $4500 = 10 x $450, a cost of $4300 = 10 x $430, and a net return of $200. Investing the money would have returned a net of $190 = 10 x $19. Because the promisee failed to understand the option nature of the contract, she wound up agreeing to a deal that left her $2.60 per contract worse off at the instant of formation."' If the promisor's return-after factoring in the actual costs, including attorney's fees-was less than that, then the promisee's failure to understand the option nature of the contract led to the formation of a contract that created a net loss to society. This result throws a wrench into claims that the option theory always maximizes social welfare. Second, unlike public law, from which individuals cannot opt out, society enforces private law only against those who express their consent to its terms through their agreement."3 Lon Fuller described the state's enforcement of contracts as "merely arming with legal sanction a rule or lex previously established by the party himself.""4 Whether one justifies contract enforcement on grounds of autonomy or welfare maximization, the parties' free choice-in other words, their consent to contract-plays a fundamental legitimating role."' The importance of such consent reappears throughout contract law. For a contract to be enforceable, the parties must have been competent to agree to it."6 Important contracts, such as those for the sale or purchase of real estate, for deals lasting more than a year, or for goods costing $500 or more, require a signed writing,'17 which among other things serves as proof of consent. A study of 187 cases involving the concept of unconscionability concluded that unconscionability could best be understood as a theory of consent."' The requirement for objective manifestations of assent," 9 not physically compelled by duress,"2 o

112. Opportunity cost minus expected gain = $19 - ($164 / 10). 113. See Barnett, supra note 43, at 320. Barnett stresses that contract law requires objective consent for practical reasons involving verifiability and induced reliance. It is thus possible for a party to be bound by a contract by which she did not want to be bound. 114. Lon L. Fuller, Considerationand Form, 41 COLUM. L. REV. 799, 806-07 (1941). 115. David Charny, Hypothetical Bargains: The Normative Structure of Contract Inter- pretation, 89 MICH. L. REV. 1815, 1848 (1991).

n6. RESTATEMENT (SECOND) OF CONTRACTS § 12 (1981) (requiring parties not to be infants, mentally ill, intoxicated, or under guardianship).

117. Id. §§ 110-150. n8. Larry A. DiMatteo & Bruce Louis Rich, A Consent Theory of Unconscionability:An Empirical Study of Law in Action, 33 FLA. ST. U. L. REV. 1067, 1115 (20o6).

11g. RESTATEMENT (SECOND) OF CONTRACTS § 18.

120. Id. § 174.

435 YALE LAW &POLICY REVIEW 30:415 2012 also ensures that the state enforces contracts only against those who consent to them. The idea of consent also plays a large role in contract interpretation. Though the parol evidence rule favors what the contract says over evidence of what the parties would have liked it to say,"' the background rule for interpret- ing anything left unclear or unaddressed in the contract is to give force to the will of the parties."' In other words, the law holds the parties to the agreement to which they consented, and if the agreement is not clear, it tries to hold them to the agreement to which they would have consented. 3 Yet American and British contract law's near-universal choice of a liability rule rather than a property rule undermines this notion of consent, at least for parties who enter into their contracts with the understanding that they are contracting for performance rather than reaching an option agreement.2 4 Most theorists ignore this point."' They see the parties' decision to contract as granting consent to be governed by the contractual rules of the jurisdiction,126 though in fact parties rarely have a choice of jurisdictions and the default rules of contract frequently conflict with unsophisticated parties' expectations about con-

121. See, e.g., Alaska N. Dev. v. Alyeska Pipeline Serv. Co., 666 P.2d 33 (Alaska 1983).

122. See, e.g., In re Columbia Gas Sys., 5o F.3d 233, 241 (3d Cir. 1995); Winet v. Price, 6 Cal. Rptr. 554> 558 (Cal. App. Ct. 1992); 4 SAMUEL WILLISTON, WILLISTON ON CONTRACTS § 6oo (4th ed. 1961) ("The guiding principle, polestar or lodestar of interpretation, whatever the form or nature of the instrument, is always the same: To ascertain the will, or intent, of the [parties].").

123. See Jules L. Coleman, Douglas D. Heckathorn & Steven M. Maser, A Bargaining Theory Approach to Default Provisions and Disclosure Rules in Contract Law, 12 HARV. J.L. & PUB. POL'Y 637, 708 (1989) (concluding that contract interpretation is governed by what these parties would have consented to rather than what objective wealth maximizers would have consented to).

124. For the classic discussion of liability rules and property rules, see Guido Calabresi & A. Douglas Melamed, Property Rules, Liability Rules, and Inalienability: One View of the Cathedral, 85 HARV. L. REV. 1089 (1972). 125. To their credit, Daniel Markovits and Alan Schwartz concede that the empirical results of Wilkinson-Ryan, Baron, Hoffman, and others pose problems for a one-size-fits-all theory of contracts; Markovits and Schwartz therefore limit their discussion to contract behavior among firms. Markovits & Schwartz, supra note 15, at 18 n.31. Schwartz's sensitivity to the different contractual preferences of different categories of contractors predates those empirical studies. See Alan Schwartz & Robert E. Scott, Contract Theory and the Limits of Contract Law, 113 YALE L.J. 541, 544-46 (2003). 126. See, e.g., KAPLOW & SHAVELL, supra note 1, at 163-64; Subha Narasimhan, Of Ex- pectations, Incomplete Contracting, and the Bargain Principle, 74 CALIF. L. REV. 1123,1179 (1986).

436 A DUAL APPROACH TO CONTRACT REMEDIES tractual obligations.2 7 If ordinary people contract for performance and the justice system delivers something else, the system "deprives them of autonomy and fails to respect their freedom to contract. "128 Perhaps that objection incorrectly assigns blame and incompletely analyzes consent. Contracts are publicly interpreted and enforced, and the public, through its government, sets the ground rules for that interpretation and enforcement." 9 Those ground rules can be read as enunciating the limits of the public's consent, such as, "We agree to enforce specific performance only when our judges determine that monetary damages would be inadequate." It follows that forcing society to deliver specific performance when monetary damages would be adequate would violate the consent of society, as expressed through its choice of contract-law doctrine. In other words, the blame for disappointing the promisee by failing to render specific performance belongs to the breaching promisor, not the nonbreaching society. But it is worth asking why contract law differs from people's expectations while other areas of the law often reflect those expectations. 30 One possible answer: Judges and lawmakers know better than the public at large which rules work best. Just as Karl Llewellyn did not merely codify the existing rules of the marketplace when he drafted the UCC but instead imposed what he considered better rules,13' Richard Craswell would not abdicate to popular promissory practice the power to determine contract law.1 2 Applying the majority's expecta-

127. See Coleman, Heckathorn &Maser, supra note 123, at 646 (pointing out that there can be no implied consent where there is no alternative to the jurisdiction's default rule); see also Randy E. Barnett, The Sound of Silence: Default Rules and Con- tractual Consent, 78 VA. L. REV. 821, 903 (1992) (arguing that "[s]ilence means consent to a particular existing default rule only when the cost of discovering and contracting around the rule is sufficiently low"). 128. Katz, supra note 29, at 788. 129. Many jurists and scholars have made the case that contract law cannot be neutral, always involves policy choices, and thus can never be purely private. See Morton J. Horwitz, The History of the Public/PrivateDistinction, 130 U. PA. L. REV. 1423, 1426 (1982).

130. For an example of public expectations' impact on the law, see Reves v. Ernst & Young, 494 U.S. 56, 66 (1990), in which the Court held that "the reasonable expectations of the investing public" trumped economic analysis in applying a test for determining what constitutes a security. In the criminal context, public expectations have framed the Court's construction of scienter requirements. See Susan L. Pilcher, Ignorance, Discretion and the Fairness of Notice: Confronting 'ApparentIn- nocence' in the CriminalLaw, 33 AM. CRIM. L. REV. 1, 4 (1996) (arguing that the "reach of the criminal law is, and ought to be, only as broad as prevailing public expectations suppose it to be").

131. Franklin G. Snyder, Clouds of Mystery: Dispelling the Realist Rhetoric of the Uni- form Commercial Code, 68 OHIo ST. L.J. 11, 13 (2007).

132. See Craswell, supra note 102, at 503, 505-o8.

437 YALE LAW & POLICY REVIEW 30:415 2012 tions to everyone still frustrates the consent of the minority, so no law can be fully consensual.'33 Not only would setting rules to match societal expectations bar progress, Craswell wrote, it also would foreclose the possibility of crafting rules that promote and reflect the society's "substantive moral values."3 4 What substantive moral value could undercompensation for breach serve? Perhaps it produces socially desirable incentives, or perhaps any benefits that full compensation creates exceed its costs. The incentives argument for undercompensation focuses on the promisee. Turning the standard justification for expectation damages on its head, consider the implications if contract law really did leave promisees indifferent between performance and damages. In such a system, the promisee would have no incentive to acquire and use knowledge about the likelihood that a promisor will perform."' Society might want to create such an incentive, because avoiding breaches reduces adjudication costs. But this approach seems flawed for two reasons. First, it penalizes those promisees who lack ex ante access to information about the likelihood that a promisor will perform. Second, because the promisor is likely to have the cheapest access to information about the likelihood of breach, the promisor should bear the costs if breach does occur."36 Contract law should incentivize honesty rather than skep- ticism. Unlike the incentives argument, which assumes that a promisee can reduce the risk of breach, the cost-benefit argument works if breach is random from the promisee's perspective. This argument concedes that, were contract law a zero-sum game, it would be wrong to redistribute wealth by taking it from nonbreaching parties, in the form of undercompensation, and distributing it to others, in the form of lower prices. But contract law is not a zero-sum game. Performed contracts increase social welfare, and fully compensatory damages probably require higher adjudication costs than undercompensatory damages. Undercompensating nonbreaching parties and lowering adjudication costs lowers prices ex ante, and those lower prices create additional gains from trade from performed contracts. Some of those additional gains from trade will go to the nonbreaching parties, and that amount might exceed the amount by which those parties were undercompensated. How could anybody object to a system in which everybody comes out ahead? But all of these assumptions are speculative. We have no evidence that current law delivers the optimal level of undercompensation, nor that the decreased damages and adjudication costs generate a sufficient drop in prices to offset the harms to nonbreaching parties. Morally,

133. See id. at 508. Note that a tailored law alleviates this issue. 134. Id. at 508-11. 135. See Samuel A. Rea, Jr., Nonpecuniary Loss and Breach of Contract, n J. LEGAL STUD. 35, 50-52 (1982). 136. See, e.g., Elliott Blanchard, Note, Terminal 25o: Federal Regulation of Airline Over- booking, 79 N.Y.U. L. REV. 1799, 1814-15 (2004) (discussing the asymmetry of information about the likelihood that a passenger will get bumped from a flight).

438 A DUAL APPROACH TO CONTRACT REMEDIES the cost-benefit argument seems to have much in common with the most con- troversial uses of eminent domain. Should the state take something from some private parties (land, or a reduction in damages) and give it to others (develop- ers, or breaching parties who do not have to pay fully compensatory damages) in the hope that the resulting private business activity will be beneficial to all (through development, job creation, and tax revenues; or through lower ex ante prices)?137 Though the benefits of undercompensation are speculative, the costs are not. Nonbreaching parties experience harms, and because promisors do not internalize the costs of those harms, promisors sometimes breach when they should be incentivized to perform. The empirical data discussed in Section II.B show that society's substantive moral values include the notion that a contract gives the promisee an entitlement greater than just the right to receive economic expectation damages in the event of breach. Many other areas of the law reject the notion that a person can compel an owner to accept court-determined damages in exchange for an own- ership right.' For example, no willingness to pay on the part of a trespasser deprives a real property owner of the right to forcibly remove the trespasser from the land.139 But the option theory of contract creates an asymmetry between the rights of sellers and buyers. Being willing to pay does not-with a few important exceptions-give a buyer the right to obtain someone else's goods or services without the prospective seller's agreement. But once a seller contracts with a buyer, the law gives the seller the right to choose whether to perform the contract or pay expectation damages, without the buyer having any power to consent to or reject the decision. Why is the nonconsensual behavior acceptable 4 ex post but not ex ante? o The nonconsensual rejection of the promisee's rights to performance brings the theory of efficient breach perilously close to endors- ing "efficient thefts."41 Those who argue for efficient breach contend that it is not nonconsensual. Shavell sees efficient breach as occurring only in response to a circumstance not covered by the contract and thus not covered by the duty to perform.142 If the

137. See, e.g., Kelo v. City of New London, 545 U.S. 469 (2005). The Kelo decision was so at odds with society's "substantive moral values," to use Craswell's expression, that forty-three states enacted post-Kelo legislation to curb the use of eminent domain. Ilya Somin, The Limits of Backlash: Assessing the PoliticalResponse to Ke- lo, 93 MINN. L. REV. 2100, 2102 (2009). 138. See, e.g., In re Indep. Serv. Orgs. Antitrust Litig., 203 F.3d 1322 (Fed. Cir. 2000) (holding that Xerox had no duty to sell or license its intellectual property). 139. See, e.g., Ariola v. Nigro, 156 N.E.2d 536, 540 (Ill. 1959) (declaring mandatory in- junctive relief to be the appropriate remedy for trespasses about which the trespasser knew or should have known).

140. Shiffrin, Breach, supra note I, at 1565-66. 141. Barnett, supra note 43, at 281 & n.46.

142. Shavell, supra note 64, at 1570-74.

439 YALE LAW & POLICY REVIEW 30 :415 2012 seller knew that the cost of performance would exceed its value to the buyer, the contract would never have been signed. Thus, when the cost-including the opportunity cost-does exceed the value, the contract should be interpreted as if the parties had considered the possibility and negotiated an economically rational term. 4 Such a term would have the parties' hypothetical consent. This hypothetical consent of the parties might be only hypothetically valid, however. The hypothetical actor reflects the views and values of her creator, which may bear no relation to those of the actual promisee. 44 People differ in risk tolerance, valuation of assets (including money), and willingness to delegate their decisions to others. Daniel Brudney describes a lottery in which there is a ninety-five percent chance that a sioo ticket will pay sio,ooo.'5 Assuming that everybody would want to buy such a ticket, he buys one for a stranger whose name he picks from a phonebook. The stranger cannot be reached before the purchase. The ticket loses. The stranger, after being told that a losing ticket was purchased for him, refuses to pay for it on the ground 1 6 that he never consented to buy it. 4 A hypothetical contract would put the promisee in a similar position to the stranger in Brudney's story; there should be no surprise if the promisee responds in a similar way.147 Imagine Brudney's story with a twist. The phone company allows people to signal their attitude about having lottery tickets purchased for them; their names will be printed in bold face if they want someone to buy them a ticket and in italics if they do not. Unfortunately, many ticket-averse people do not know that they need to signal their choice, and ticket purchasers typically ignore the typefaces. In spite of the signaling mechanism, mismatches in this "market" persist. This condition describes the current dysfunctional state of contract law. Parties can signal whether they want to contract for performance or contract for value by setting the appropriate level of liquidated damages. A high value can be seen as emphasizing that performance is expected; a lower

143. Id. 144. Victor Brudney, Contract and Fiduciary Duty in Corporate Law, 38 B.C. L. REV. 595, 637 n.112 (1997) ("The hypothetical actor is simply a fictitious implementer of the contractarians' conception of an efficient solution."). In a similar vein, Jules Coleman noted that "there appears to be nothing expressed by the concept of hypothetical consent that is not already captured in the idea of rational self-interest." JULES L. COLEMAN, RISKS AND WRONGS 169 (1992).

145. Daniel Brudney, Hypothetical Consentand Moral Force, io LAW & PHIL. 235, 238-39 (1991).

146. Id.

147. The situations are not identical, in that hypothetical contracts "complete" existing contracts instead of creating new contractual relationships. But just as Brudney imposes on the stranger certain views about acceptable risk and rationality, a judge, when creating a hypothetical contract, imposes on the promisee certain views about acceptable risk and rationality.

440 A DUAL APPROACH TO CONTRACT REMEDIES value can be seen as suggesting an option contract.' 4 But this solution has three flaws. First, those who desire to contract for performance often lack the sophistication to realize that they need to signal that desire. If contracting for performance is not the default, then unsophisticated parties who prefer a performance contract will still wind up in option contracts.149 Second, courts often refuse to enforce high-value liquidated damages on the grounds that public policy re- stricts the remedy for nontortious contract breach to compensatory damages. 5 o The same courts that refuse to pass judgment on the appropriateness of a contract's price term will refuse to enforce a liquidated damages term that they deem punitive."' Third, courts will refuse to enforce liquidated damages unless actual damages were sufficiently uncertain at formation.1'5 Similarly, courts may refuse to enforce a contract that calls for specific performance in the event of breach.'53

D. The Search for a Middle Ground

Richard Brooks rejects the choice between the option theory and the performance theory.5 4 He seeks the benefits of both theories through the estab- lishment of a middle ground that preserves the economic efficiencies of breach without assuming the promisee's indifference to it. Brooks points out that contractual benefits include rights as well as rents and that there is no fixed conver- sion rate between the two-no contract theory equivalent of what Einstein's

148. Schwartz asserts that courts should respect liquidated damages as accurate reflec- tions of how parties value performance. Though he does not distinguish between the option and the performance theory, his approach would accommodate those who contract for performance. See Schwartz, supra note 33, at 383-87. 149. See Shiffrin, Breach, supra note i, at 1561 (stating that allowing parties to declare that they desire an option contract does not solve the larger issue of where to set the default rule). 150. See Scott & Triantis, supra note 59, at 1429, 1461-62, 1467, 1476-77 (arguing that "the right to breach and pay damages is ... a valuable option sold by the promisee to the promisor" and that courts therefore should respect the valuation of that option set by the parties as liquidated damages, even when such damages seem supercompensatory).

151. See Schwartz, supra note 33, at 370 (calling for abandonment of the ex ante branch of the liquidated damage rule because courts sometimes mistake compensatory damages clauses for penalties and refuse to enforce them).

152. RESTATEMENT (SECOND) OF CONTRACTS § 356 (1981); U.C.C. § 2-718. 153. See supra notes 27-31 and accompanying text.

154. See Richard R.W. Brooks, Participation Interests (Feb. 20, 2011) (unpublished manuscript) (on file with author).

441 YALE LAW& POLICY REVIEW 30:415 2012

E=mc' did for energy and mass."' Parties are likely to view forced conversions of rights into rents as unjust. Brooks illustrates this tendency with Shiffrin's story of the no-show plumber.'56 The plumber, without prior negotiation or con- sultation, takes it upon herself to convert the buyer's right to service-as-scheduled into a reduction in rent. In Brooks's view, what the customer really wants, short of specific performance, cannot adequately be captured in monetary terms. This "participation interest" might be satisfied with something as simple as a phone call or a text message, or it might be eliminated altogether through a contract that explicitly outlines the possibility of efficient breach and spells out its implications.' Wilkinson-Ryan and Baron provided empirical support for this notion when they asked survey respondents to calculate damages owed by a restaurant owner who breaches a contract to host a couple's anniversary party."' Damages were almost the same whether calculated at formation of the contract or after a phone call from the owner to the couple explaining his decision to breach and his reason for doing so. Damages were higher, however, if the couple got no chance to negotiate them and the amount was instead set by an impartial medi- ator.59 Wilkinson-Ryan and Baron described this outcome as a timing effect, with higher damages after the scheduled time of performance than before it; however, their data support the interpretation that the key difference is whether the promisees get to participate in the damages-setting process. Economic efficiency does not require that society assign to the promisor either the rent from efficient breach or the right to choose efficient breach.' Brooks shows that a promisee's option to force performance or to choose non- performance with disgorgement produces the same level of efficiency as giving both the rent and the right to the promisor. The best solution, he suggests, might come from considering a third factor along with the rents and rights analysis: power. "The promisor retains the power over the choice to perform the promised deed, but not the right."' This efficient performance theory may be viewed as a middle ground between the extreme versions of the option theory, where the promisee has no say, and the performance theory, where the contract gives the promisee an entitlement to the subject matter of the promise.

155. The "rents" here are economic rents, which are "[t]he return[s] gained from an economic resource (such as a worker or land) above the minimum cost of keeping the resource in service." BLACK's LAW DICTIONARY 590 (9 th ed. 2009). 156. Brooks, supra note 154, at 8-9 (quoting Shiffrin, Breach, supra note I, at 1564). 157. Id. at io & n.14, 11. 158. Wilkinson-Ryan & Baron, supra note 68, at 415-17. 159. Id. 160. See Brooks, supra note 93. 161. Id. at 593.

442 A DUAL APPROACH TO CONTRACT REMEDIES

Perhaps a waiver of participation interest is built into the market price.162 The promisee's payoff for waiving any participation interest in cases of efficient breach would be the lower price at which the other party can-and must-be willing to contract in exchange for not having to share the benefits of breach or bear the expenses of notification. This theory encompasses two possibilities: (i) The default state is an example of the market giving promisees what they want, so any complaints they have about a denial of their participation interest are hypocritical, and (2) the default state is an example of the market's failure to produce outcomes that promisees would prefer. The first possibility echoes the hypothetical consent notion discussed in 6 Section C and appears similarly unpersuasive.' 3 The truth is that parties are not generally given the choice of obtaining a discount in exchange for relinquishing their performance interest. Even if consumers had been given such a choice, it would be impossible to say simply that the consumers got the tradeoff between the price and participation interest that they would prefer. Preferences are different at formation than at the time of performance or breach. 6 4 This difference will be especially pronounced for unsophisticated contractors experiencing breach for the first time, who belong to the category of litigants Marc Galanter 6 refers to as one-shotters. 5 Optimism bias leads one-shotters to undervalue, at the time of contracting, all terms related to breach,' 6 so, if given the choice, they would typically show a willingness to sacrifice the participation interest too cheaply in the quest for a lower price.117 In contrast, repeat players who contract with the same party on multiple occasions may acquire enough information to demand the correct discount for waiving their performance interest.

162. See Katz, supra note 29, at 782-83. 163. See supra text accompanying notes 128-146. 164. See Melvin Aron Eisenberg, The Limits of Cognition and the Limits of Contract, 47 STAN. L. REV. 211 (1995). 165. See Marc Galanter, Why the 'Haves' Come Out Ahead: Speculations on the Limits of Legal Change, 9 LAW & Soc'y REV. 95, 97 (1974). 166. See Eisenberg, supra note 164, at 216-18.

167. Id. at 227 (referring to parties' difficulties in imagining breach); cf Barnett, supra note 30, at 181 (stating that parties who expect specific performance as a remedy will not get the price they should when bargaining with parties that know the risk of enforcement error). A well-functioning market would be characterized by parties contracting as they would if they had complete knowledge-and comprehension-of the likelihoods of possible outcomes. But a well-functioning market would not necessarily choose the outcome that maximizes expectation-the sum of the net benefit of each outcome multiplied by its probability. The one-shotter is likely to be sub- stantially more risk-averse than the repeat player and thus would be willing to pay a premium for an increased assurance of performance. Giving the one-shotter a price discount in exchange for the participation interest runs counter to that idea.

443 YALE LAW& POLICY REVIEW 30:415 2012

The idea of a participation interest achieves the important objective of giving a name to the gap between the damages to which a promisee is entitled and the promisee's true value of performance. The lack of equivalence between damages and performance must be addressed if contract law is to be grounded in the consent of the parties. Courts generally have failed to address this gap.

III. How COURTS TACKLE THE QUESTION

Both common law and statutory rules tend to reflect an option theory of contract, as discussed in Section I.B, but exceptions exist, as shown by the disgorgement remedy also noted in Section I.B and Galanter's research discussed in Section II.B.'" Tensions between performance-based and option-based views have surfaced, for example, in cases involving bad-faith breach and the choice of damages between cost of completion and diminution in value.

A. Bad-Faith Breach

A purely option-based view of contract treats breach with damages as equivalent to performance; the promisor chooses, the payment occurs, and neither the promisee nor the court deserves an explanation. In practice, the law is not quite that simple. As of 1999, at least a dozen states allowed or appeared to allow punitive damages for breach of contract, under limited circumstances, without proof of an independent tort. ' Judge Posner supports the imposition of punitive damages in cases of opportunistic breach, which William Dodge defines as including pretextual termination, stonewalling, and bad-faith refusal to pay a debt.' Be- cause these breaches make no claim to economic efficiency-at least when viewed from the perspective of society as a whole rather than the breacher- even option theorists see no grounds on which to encourage them.

168. See supra text accompanying notes 40-41 (disgorgement) and 78-81 (Galanter's research). 169. William S. Dodge, The Case for Punitive Damages in Contracts, 48 DUKE L.J. 629, 647 (1999). Four of those states allowed such damages only where there was a preexisting special relationship between promisor and promisee. A special relationship in this context is one that creates a duty that extends beyond the terms of the contract, such as the duties between "insurer and insured, carrier and passenger, innkeeper and guest, physician and patient, and attorney and client." Id.

170. Id. at 656 (citing RICHARD A. POSNER, EcONOMic ANALYSIS OF LAW 142 (5th ed. 1998)). Stonewalling in the contractual context refers to taking a "see you in court" approach in an attempt to avoid liability where no defense exists. See Sea- man's Direct Buying Serv. v. Standard Oil Co., 686 P.2d 1158, 1167 (Cal. 1984) (per curiam), overruled by Freeman & Mills, Inc. v. Belcher Oil Co., 900 P.2d 669 (Cal. 1995).

444 A DUAL APPROACH TO CONTRACT REMEDIES

These bad-faith breaches have something else in common, too: They evoke moral outrage, a response that performance theorists and legal realists might say explains the availability of punitive damages. In Seaman's Direct Buying Ser- vice v. Standard Oil Co., a ship-supplies dealer needed the defendant to deliver on its contract, which was a prerequisite for maintaining the plaintiffs marina lease. 7 ' The defendant tried to get out of its contractual obligation and eventu- ally simply denied that a contract existed. This breach forced the plaintiff out of business. The California Supreme Court analogized the defendant's stonewalling to threatening to bring a lawsuit in bad faith, an action for which punitive damages would be available, and thus held that punitive damages are available for bad-faith denial that a contract exists. 72 But courts have also applied punitive or emotional distress damages in cases that could be classified as examples of efficient breach. In Dold v. Outrigger Hotel,"' the Supreme Court of Hawaii approved of a jury instruction allowing damages based on the nonbreaching party's emotional distress and disap- pointment. Two couples had booked rooms at the Outrigger but were trans- ferred to lesser-quality accommodations. Similar fates had befallen other guests; the Outrigger consistently overbooked its rooms and then profited by pocketing the difference between the amount it had charged for the rooms and the amount it paid for rooms at the overflow hotel. The plaintiffs won on the merits at trial but appealed the judge's refusal to issue a jury instruction allowing punitive damages. The resulting affirmance opined approvingly about the jury instruction the judge did give, allowing emotional distress damages: "[C]ertain situations are so disposed as to present a fusion of the doctrines of tort and contract."7 4 That personal contract dictum became commercial contract law in Chung v. Kaonohi Center Co.17 The plaintiffs entered a contract for a ten-year lease to operate a fast-food outlet in a mall that the defendants were building. The plaintiffs put down a deposit, arranged "financing, ordered equipment and fur- nishings, hired chefs and workers, advertised in the yellow pages ... and incurred other expenses.""' The plaintiffs and the defendants were in frequent contact, with plaintiffs expressing their eagerness to know when they could open their outlet.'7 Meanwhile, the defendants were negotiating with other par-

171. 686 P.2d 1158.

172. Id. at n67.

173. 501 P.2d 368 (Haw. 1972).

174. Id. at 372. 175. 618 P.2d 283 (Haw. 1980), overruled by Francis v. Lee Enters., 971 P.2d 707, 712-15 (Haw. 1999).

176. Id. at 286.

177. Id.

445 YALE LAW & POLICY REVIEW 30 :415 2012 ties for the same location and gave a right of first refusal to one of those parties, all without informing the plaintiffs.'17 When the plaintiffs saw an article naming one of the other parties as the lessee, one of the defendants denied the newspaper's report.179 Finally, a year after the lease was signed, the same defendant returned the deposit and informed the plaintiffs they would not get the lease.so The Chung court upheld an award of damages for emotional distress based on "wanton and/or reckless conduct sufficient to give rise to tort liability."'' Aside from the dishonest response about the newspaper article, Chung looks like a straightforward case of efficient breach. The option theory would argue that expectation damages would have been sufficient, with no recovery for emotional distress. Courts increasingly see things as the option theory does. The trend has been away from punitive damages, other than in the context of bad-faith denials of insurance claims or breaches with an accompanying action in tort. In 1999, the Hawaii Supreme Court abolished the Dold-Chung rule allowing tort liability for willful, wanton, or reckless contract breach.' The court rea- soned that unlike tort law, which vindicates social policy, contract law is about effectuating the parties' intentions.' The court went on to say that the intentions of the parties to a contract include foreseeability of damages.184 As explained in Section I.B, this logic indirectly refers to the option theory. Four years earlier, the California Supreme Court also listed foreseeability as a reason to overturn Seaman's, which it said had opened a Pandora's box of confusion by applying tort damages to a non-insurance case of bad-faith breach.'* In Freeman & Mills, Inc. v. Belcher Oil Co., the court rejected the idea that stonewalling worthy of punitive damages was qualitatively distinguishable from actions that "could incidentally accompany every breach of contract."' In dissent, Justice Mosk argued that punitive damages should be available for breach when the breach is accompanied by a tort or is accomplished by tortious means "involving deceit or undue coercion," or where "one party intentionally breaches the contract intending or knowing that such a breach will cause severe,

178. Id.

179. Id. 180. Id. 181. Id. at 289. 182. Francis v. Lee Enters., 971 P.2d 707, 712-15 (Haw. 1999).

183. Id. at 712.

184. Id. at 712-13. 185. Freeman & Mills, Inc. v. Belcher Oil Co., 900 P.2d 669, 674, 680 (Cal. 1995) (citing foreseeability as a reason to overrule Seaman's); id. at 675-76 (citing confusion as a reason to overrule Seaman's). 186. Id.

446 A DUAL APPROACH TO CONTRACT REMEDIES unmitigatable harm in the form of mental anguish, personal hardship, or substantial consequential damages."' Those reasons explicitly link Justice Mosk to the performance theory's recognition that many breaches leave the promisee far from indifferent between performance and expectation damages. Even the majority hinted that it understood that point with its suggestion that the legislature might wish to consider "providing litigation costs and attorney fees in certain aggravated cases, or assessing increased compensatory damages covering lost profits and other losses attributable to the breach."' 8

B. Diminution in Value vs. Cost of Completion

If parties contract for value, then the remedy for breach should be limited to the resulting diminution in value. If they contract for performance, the remedy should be either specific performance or a sufficiently large damages award to allow the promisee to obtain the promised performance. Thus, cases involving a choice between diminution in value and cost of completion provide in- sights into the value versus performance debate. On the value side stands Peevyhouse v. Garland Coal & Mining Co.,' 9 in which the Oklahoma Supreme Court reduced damages against a coal company that had intentionally breached its agreement to restore a family's farm after concluding its strip mining operation. The remedial work the contract called for would have cost $29,000; the farm would have been worth less than $5,000 after that work was done.'90 The court refused to enforce the contracted-for performance on the grounds that "the economic benefit of performance would be grossly disproportionate to its cost" and instead set the damages at $300, the diminution in the farm's value caused by the company's failure to do the remedial work.' 91 The family surely would have negotiated differently if it had known it was contracting for the value of the remedial work rather than for the remedial work itself. And at the level of damages as measured by the court, the family surely was not indifferent between receiving performance and receiving $300. By not taking into account the subjective value of performance, and the objective fact that the family negotiated for remediation of the land (not a $300 increase in the land's value), the option approach left the family undercompensated. In contrast, in American Standard, Inc. v. Schectman,192 the New York Su- preme Court, Appellate Division, followed the performance theory. Unlike an

187. Id. at 681 (Mosk, J., dissenting). 188. Id. at 68o (majority opinion). 189. 382 P.2d 109 (Okla. 1962). 190. Id. at in. 191. Id. at 114.

192. 439 N.Y.S.2d 529 (App. Div. 1981).

447 YALE LAW& POLICY REVIEW 30:415 2012 option-theory court, it cared about the reason for breach, holding that Schectman's failure to perform the contract was willful and that diminution in value was available as a measure of damages only in cases in which the promisor showed substantial performance made in good faith.'93 In other words, diminution in value as a measure of damages was not an option a promisor could select before breach but only a remedy that might be applicable post-breach. Both the Peevyhouse and the American Standard courts noted a problem with the option theory, at least as measured by diminution in value. If a person contracts for work that will decrease the objective value of her property, then under the diminution-in-value measure a breach would be cost-free to the breacher. Economically, welfare maximizers could argue, society would be better off that way. A nineteenth-century New York court ruled otherwise: A man may do what he will with his own,.. . and if he chooses to erect a monument to his caprice or folly on his premises, and employs and pays another to do it, it does not lie with a defendant who has been so employed and paid for building it, to say that his own performance would not be beneficial to the plaintiff.194 The contract can still be seen as granting an option between performance and the objective value of the work, rather than the objective value of the work's results. But it also can be seen as putting the promisee in a position to collect subjective, performance-centered damages, as there is no objective economic value lost in the event of breach.

IV. POTENTIAL SOLUTIONS: WHY BIFURCATING REMEDIAL APPROACHES MAKES SENSE

There are multiple solutions for addressing the difference in expectations arising from the disconnect between the performance view of contract that most people hold and the option view of contract favored by the law and sophisticated parties. We can decide to do nothing, either because we deny the problem exists or consider it insoluble. We can educate people about the option theory so that they understand what they are consenting to when they enter a legally enforceable agreement. We can alter the way the legal system enforces contracts so as to make it match unsophisticated people's performance-oriented

193. The defendant failed to complete grading and removal of subsurface structures on the plaintiffs property. The court refused to hear testimony that the plaintiff had sold the land for only $3000 less than its fair market value. The Appellate Division upheld the trial court's rejection of the testimony and affirmed the jury's $90,ooo verdict. Id. at 530-31.

194. Chamberlain v. Parker, 45 N.Y. 569, 572 (1871).

448 A DUAL APPROACH TO CONTRACT REMEDIES theory. Or we can create dual systems that match the expectations and needs of the parties.' 5

A. Doing Nothing

Doing nothing undermines the idea that public enforcement of private law is both just and fitting only when parties have formed that private law consen- sually. Contract law cannot be consent-based if the parties do not know what they have consented to, and that is far too often the case. Evidence shows that most people hold a performance-based view of contract but that courts generally enforce contracts as options. In light of this, the only excuse for inaction would be that there is no action that could work. The rest of this Part illustrates why that is not the case.

B. Notice

If people are merely misinformed about the way contracts work, the simplest solution might be to educate them. The option theory of contract could become a required element in every high-school civics class. Promisors who want to retain the option to breach and pay only economic damages could be required to state that option explicitly in the contract, thereby replacing the hypothetical consent currently ascribed to promisees with actual written consent. More generally, contracts could be unenforceable without a state-mandated explanation of the way in which the judicial system responds to breach. Shiffrin mentions something akin to this idea in passing, but does not explore the details, which are not straightforward.296 For instance, a warning in a contract might well go unread, even if each party were required to initial or sign the document to signal both reading and comprehension of the specified language.19 7 Perhaps there could be a requirement that the words be read aloud as a sort of Miranda warning; instead of learning their rights, signatories would learn their lack of rights to receive the contracted-for performance rather than damages determined by a court-or, more likely these days, an arbitrator. This

195. Another possibility is that we already have dual systems but never say so. Juries and judges may already apply the law in different ways depending on the status of the parties. See infra Section IV.D. 196. Shiffrin, Divergence, supra note i, at 742 ("[W]e could identify the onset of contractual relations in a clearer way to put parties on greater notice that these agreements are subject to special rules and may be treated differently than promises."). 197. See Omri Ben-Shahar, The Myth of the 'Opportunity To Read' in Contract Law, 5 EUR. REV. CONT. L. 1, 2-3 (2009) (arguing that, "[alpart from an exotic individual here or there, nobody reads" standard-form contracts and that it is "counter-realistic" to build a theory of consent on the notion that people will).

449 YALE LAW & POLICY REVIEW 30:415 2012 approach would create more informed consent, at the cost of adding another step to the formation of every contract. But just as crime suspects who tell offic- ers that they understand their Miranda rights routinely say things that suggest they do not really understand, promisees who have been warned that they have signed an option contract rather than a performance contract might still consider themselves entitled to performance.'9' People often prove difficult to educate. Even if notice worked as intended, the result still would be a system of contract law inconsistent with the rules that studies show most people would prefer. Fixing the consent problem through education does not alter the disparity between the rules that exist and the ones that would best fulfill parties' needs and expectations. People wishing to contract for performance would face the option of contracting for value or not contracting at all. Notice therefore supplies a relatively cheap but decidedly imperfect solution to the consent problem.

C. Remedies Under a Performance Theory

If notice bridges the gap imperfectly by bringing people's understandings into harmony with the law, a better alternative might be to bring the law into harmony with people's understandings. The fit might not be perfectible, as some of people's understandings about contracts might prove unworkable when codified or might conflict with more important societal values. But there is no reason not to do the best we can. Ideally, a performance-based theory would grant specific performance when the plaintiff desired, provided that such a remedy would be consistent with limitations imposed by principles of inalienability and pragmatism. Oth- erwise, a performance-based theory would grant damages that reflected not only the objective value of performance but also the nonpecuniary harm from breach. Some degree of fee shifting would be available for victorious plaintiffs, although forcing plaintiffs to continue to bear a certain level of litigation costs would be necessary as a motivation for settlement. 99 One way to conceptualize damages rules under a performance-based theory would be to think of a performance-based contract as consisting of two option-based ones. The first would be the contract for the good or service itself.

198. E.g., WELSH S. WHITE, MIRANDA'S WANING PROTECTIONS 166 (2003) (quoting an exonerated suspect as saying that he gave the false confession that led to his con- viction because he wanted to end the interrogation); Susan R. Klein, Identifying and (Re)formulatingProphylactic Rules, Safe Harbors,and Incidental Rights in Con- stitutional CriminalProcedure, 99 MICH. L. REV. 1030, 1071 (2001) (concluding that "Miranda is relatively ineffective at dispelling coercion"); see also Shiffrin, Diver- gence, supra note I, at 743. 199. See Schwartz, supra note 33, at 396 (suggesting that fee shifting is a logical contract term that a promisee would desire, because anything more than full expectation damages plus legal costs would be economically inefficient).

450 A DUAL APPROACH TO CONTRACT REMEDIES

The second would be the contract for the metaconcept of performance, which may also be thought of as the value of keeping a promise. That value equals compensation for the frustration that the promisee experiences because the remedy is not specific performance, delivered at the promised time and without any of the annoyances and distress that come with breach. Contract law currently recognizes the first contract and ignores the metacontract, but the harm from breach of the promise to perform remains real to the promisee. Shiffrin notes that "alterling] the significance and appropriateness of others' reactions to a broken obligation is not within the power of the promisor;"2oo neither is it within the power of the system of contract law. The promisee's reaction to the broken obligation depends at least partly on the reasons for and the timing of breach. Damages from breach of the metacontract would be monetizable, possibly through legislatively created guidelines similar to sentencing guidelines. Perhaps one would start with a baseline multiplier, say five percent of the first contract's price, and adjust the figure up or down based on circumstances, such as the time between repudiation and scheduled performance, the reason for breach, and the availability of mitigation. Avery Katz suggests compensating the promisee by an amount sufficient to offset the "disutility associated with the mental adjustment of receiving" something other than the contracted-for performance.2 0 ' Though litigating those details would drain court resources and increase the parties' legal costs, those negatives would be at least somewhat offset by decreases in breach and increases in orders of specific performance. Breach would decline as it became more costly to promisors; specific performance orders relieve the court of the burden of calculating damages. Inevitably, some promisees will suffer harms far beyond those experienced by the average victim of breach. A tort-like system would fully compensate those promisees under the principle that you take your victim as you find her.2 0 2 But such an approach does not comport with the realities of contract law, where breach does not imply culpability and the "victim" of a breach chose at formation to form a relationship with the promisor. It therefore might make sense to cap any emotional distress or annoyance damages at an amount that is reasonable under the circumstances of the breach and foreseeable at the time of breach.2 0 3 The "reasonable under the circumstances of the breach" limitation could be a means of preventing a promisee

200. Shiffrin, Divergence, supra note i, at 729.

201. Katz, supra note 29, at 784.

202. Vosburg v. Putney, 5o N.W. 403 (Wis. 1891), offers the classic example of this rule's application. 203. For a discussion of why foreseeability at breach would be preferable to foreseeability at formation as a limitation under a performance-oriented theory of contract, see supra note 35 and accompanying text.

451 YALE LAW & POLICY REVIEW 30 :415 2012

from overreliance on performance, a problem discussed in Section I.B.20 4 Both the reasonableness and the foreseeability limitations would provide incentives for the parties to keep one another informed about circumstances relating to the contract. The promisor would provide the promisee early notification of problems meriting a decrease in reliance on performance. 0 The promisee would provide the promisor with information about the extent of her reliance. Just as in the option approach, damages would be set at levels that force the parties to internalize the costs of their actions. The differences would arise from the definition of what constitutes a compensable harm. Damages for breach of the metacontract-the failure to deliver on an obligation-should thus be set to send the appropriate price signal, but, unlike damages in traditional contract remedies, that price signal would not be independent of the reason for breach. Because the fairness of a breach affects the harm felt, it should affect the damages assessed. Putting a price on fairness seems like a foreign concept to contract law, but it can be done. Consider, for example, the ultimatum game, in which a propos- er suggests an allocation of a sum of money, and a decider chooses either to accept the allocation, in which case both parties get paid as proposed, or to reject the allocation, in which case neither party gets paid.2"' The decider's willingness to forgo money in order to express the unfairness of a proposed allocation shows that the price she puts on that level of unfairness is as high as the money she would have received. But a performance-based approach to remedies would not fit everybody's needs. Those who contract for value would have no desire for specific performance and no interest in expending resources to determine damages based on the nonbreaching party's subjective valuation of performance. For businesspeople, current law offers the more efficient alternative.

204. See supra note 36 and accompanying text. Though courts would find it difficult to measure the optimal level of reliance, see Eisenberg & McDonnell, supra note 36, at 1371, a "fair and reasonable approximation" suffices for calculating damages, Nat'l Austl. Bank v. United States, 452 F.3d 1321, 1327 (Fed. Cir. 2006). 205. Craswell has suggested optimizing investment on reliance by forcing a similar kind of communication by the promisor to the promisee, but in his analysis the communication occurs at formation. Richard Craswell, Performance, Reliance, and One-Sided Information, 18 J. LEGAL STUD. 365, 367-68 (1989). If notifications about the likelihood of performance come after formation, the notifications would need to be backed by some kind of evidence; a promisor without concern for her reputation would have an incentive to cry wolf about the possibility of breach and thereby lower her exposure to reliance-based damages. Alternatively, the law could impose liability on a promisor for misrepresenting the likelihood of breach. See id. at 375-76. 206. See Brooks, supra note 154, at 13.

452 A DUAL APPROACH TO CONTRACT REMEDIES

D. CreatingDual Systems

The best solution is to assess remedies based on the contract theory that best suits the parties' expressed or probable desires. A bifurcated system of contract remedies gives the economic power of the option theory to those who contract for value and a more fully compensatory selection of remedies to those who contract for performance. Establishing such a dual system would be consistent with other areas of contract law, would increase the law's legitimacy, and would be workable. Subsection IV.D.1 shows that contract law is not and need not be one-size-fits-all. Subsection IV.D.2 explores the issue of which contracts should be governed under the option theory and which should be governed under the performance theory. Subsection IV.D.3 analyzes the costs and benefits of making the choice of contract theory contractible.

1. How Contract Law Already Tailors Itself to Parties' Needs

Arguing that contract law should grant different remedies for breach based on what the "parties would tend to prefer" is hardly revolutionary.2 0 7 Tradition- ally, though, the law locates that preference in the nature of the contracted-for performance rather than in the status of the parties.208 But there is no reason this must be so.09 For example, Ian Ayres once floated the idea of giving consumers more fully compensatory damages than businesses." 0 Contract law already establishes different rules for different people. Most famously, Article 2 of the UCC recognizes that the needs of merchants differ

207. See Steven Shavell, Specific PerformanceVersus Damagesfor Breach of Contract: An Economic Analysis, 84 TEx. L. REV. 831, 832 (2006). 208. See, e.g., id. (urging the imposition of specific performance of contracts for con- veyance of property, not just land, and the imposition of damages for breaches of contracts to produce things). 209. Daniel Markovits identifies three distinct categories of contracts: those "between individuals,... between organizations, and ... between individuals and organizations," and he asks whether there should be one law of contracts for the first category and another for the latter two categories. Daniel Markovits, Contract and Collaboration, 113 YALE L.J. 1417, 1464-65 (2004). Ethan Leib asks why Markovits did not suggest three laws of contracts, one for each category. Ethan J. Leib, On Collaboration, Organizations, and Conciliation in the General Theory of Contract, 24 QUINNIPIAc L. REV. 1, 12 (2005). Alan Schwartz and Robert Scott add a fourth category by distinguishing between contracts in which a firm sells to an individual and contracts in which an individual sells to a firm. Schwartz & Scott, supra note 125, at 544. 210. Ian Ayres, Preliminary Thoughts on Optimal Tailoring of Contractual Rules, 3 S. CAL. INTERDISC. L.J. 1, 4 n.15 (1993) (suggesting, as an example of a tailored default rule, giving reliance damages for breaches of wholesale contracts and expectation damages for breaches of retail contracts).

453 YALE LAW & POLICY REVIEW 30 :415 2012

2 11 from those of less-sophisticated parties to sales contracts. Article 2 establishes fourteen situations in which there is one set of rules for merchants and another for all other parties.2 12 Sometimes the rules lower protections for merchants, thereby allowing more efficient transactions. For example, an exception to the statute of frauds allows the necessary writing to be produced within a reasonable time after formation when the deal is among merchants. 13 Sometimes the rules raise protections for those doing business with merchants. For example, merchants' firm offers are not revocable without consideration.2 14 And sometimes the rules both lower protections for merchants and raise protections for nonmerchants. For example, merchants can more easily adjust the terms of a contract among themselves,"' but nonmerchants get protection from "no oral modification" clauses; if a merchant includes such a clause in a form contract, the clause is invalid against a nonmerchant if the nonmerchant has not signed a separate agreement to that clause."' The general idea is that the laws necessary to protect unsophisticated contractors make no sense for sophisticated businesspeople. 17 The American Arbitration Association applies this principle to its rules, too. Its Consumer Due Process Protocol sets a minimum standard for arbitration clauses in consumer contracts."' This minimum standard does not apply to contracts among big businesses, which presumably can fend for themselves, but it might apply by analogy to small businesses and individual employment contracts."9 The protocol sets default rules, not immutable rules; consumers may waive the protections "[a]ssuming they have sufficient specific knowledge and understanding of the rights they are waiving."2 2 0 Another protocol covering employee arbitration "has won broad judicial and academic acceptance and in- forms the drafting of arbitration policies satisfying due process concerns in al-

211. See, e.g., U.C.C. §§ 2-201(2), 2-205, 2-207(2), 2-209(2), 2-605(1)(b), 2-609(2) (1977). 212. Ingrid Michelsen Hillinger, The Article 2 Merchant Rules: Karl Llewellyn's Attempt To Achieve the Good, the True, the Beautiful in Commercial Law, 73 GEO. L.J. 1141, 1143 & .111(1985). 213. U.C.C. § 2-201(2). 214. Id. § 2-205. 215. Id. § 2-207(2). 216. Id. § 2-209(3). 217. Hillinger, supra note 212, at 1147-48.

218. See AM. ARBITRATION Ass'N, CONSUMER DUE PROCESS PROTOCOL (1998), available at http://www.adr.org/aaa/ShowPDF?doc=ADRSTG-oo5o4. 219. Id. ("Scope of the Consumer Due Process Protocol"). 220. Id. princ. i cmts.

454 A DUAL APPROACH TO CONTRACT REMEDIES most every, if not actually every, jurisdiction."2 2 1 But though the American Ar- bitration Association may decline to arbitrate when arbitration clauses do not meet the protocols' requirements,2 2 2 it does not have exclusive jurisdiction, and businesses are free to have their arbitrations performed under the auspices of a national association with fewer consumer-protection scruples.22 3 Still, the existence of these protocols shows the widespread recognition that, in dispute reso- lution, different kinds of parties require different kinds of background rules. A party's ability to enter into a contract often varies based on the party's status and the type of contract. For example, Securities and Exchange Commis- sion regulations differentiate between unsophisticated individuals and "accred- ited investors" who are eligible to purchase certain high-risk investments.'2 Be- coming an "emancipated minor" enables a person under age eighteen to enter into contracts."' Only a homeowner age sixty-two or older, or one with a spouse that age, can enter into a contract for a reverse mortgage on her house."' Though neither statutory nor common law calls for a party's status to affect the choice or extent of contract remedies, such an effect might already exist in practice. Juries might vary damages awards based on the status of the nonbreaching party and the nature of the breach, rather than complying with the judges' instructions.2 7 But even if such de facto bifurcation of the rules governing remedies is taking place, making the bifurcation explicit would improve contract law by increasing predictability. Those who contract for value would continue to be governed by current law. Those who contract for performance would receive specific performance where it is desired, pragmatic, and consistent with public policy, and would otherwise receive fully compensatory damages. By allowing the minority to follow the rules it deems economically ef-

221. David Sherwyn, Samuel Estreicher & Michael Heise, Assessing the Case for Em- ployment Arbitration: A New Path for Empirical Research, 57 STAN. L. REV. 1557, 1562 (2005). 222. W. Mark C. Weidemaier, Arbitration and the Individuation Critique, 49 ARIz. L. REV. 69, 87-88 (2007). 223. See Stephen K. Huber, State Regulation of Arbitration Proceedings: Judicial Review of Arbitration Awards by State Courts, io CARDOZO ONLINE J. CONFLICT RESOL. 509, 554 (2009) (describing the National Arbitration Forum as "more favorable to merchants").

224. See 17 C.F.R. §§ 230.501-.508 (2012).

225. See Carol Sanger & Eleanor Willemsen, Minor Changes: Emancipating Children in Modern Times, 25 U. MICH. J.L. REFORM 239, 240-41 (1992).

226. 12 U.S.C. § 17 15 z-20(b)(1) (2006). 227. For a description of jury behavior in contracts cases, see supra text accompanying note 81. For a description of potential jurors' sense of justice in contract cases, see supra text accompanying notes 68-76.

455 YALE LAW & POLICY REVIEW 30:415 2012 ficient and the majority to follow the rules that best comport with its expectations, all parties benefit.

2. Where To Apply the Option and Performance Theories

Multiple possibilities exist for deciding which parties' contracts will fall under performance-oriented remedy rules and which will fall under option-oriented remedy rules. The division could be set to coincide with the UCC definition of "merchant," though that definition has not proven particularly clear in practice.228 The default rule could apply the option theory only to contracts among merchants, or only to contracts in which a merchant is the promisee, with the idea being that only nonmerchant promisees would want and need the protections that a performance-based rule would provide. Intuition suggests that some merchants belong on the performance side of the performance/option divide and that even many contracts between businesses are not intended as option contracts. For example, though the bad-faith breach case of Chung, discussed in Section III.A, involved a contract between two partnerships, the court nonetheless upheld an award of damages for emotional distress.12 9 The court rejected the defendants' contention that, because a business contract's primary objective is financial gain, emotional-distress damages were unforeseeable and thus inappropriate.23 o Though businesses cannot experience emotional distress, the people who run businesses can. As Alan Schwartz and Robert Scott argued, the contract-law needs of a gift shop run by a retired teacher might have more in common with those of ordinary consumers than with the needs of publicly held corporations. 3' The latter want to be governed by contract law focused exclusively on promoting efficient trade and investment,2 32 so laws consistent with the option theory ought to apply to their dealings with other big firms. Schwartz and Scott suggest applying efficiency-centered rules to companies of five or more employees, limited partnerships, and professional partnerships.2 3 But having one rule for one- to four-employee businesses and another for larger ones would create two serious problems. The first is economic: Small businesses compete with larger ones, and if the larger ones operate under more efficient contract rules they will receive a competitive advantage. Consider a book publisher contracting with bookstores of different sizes. It will charge

228. Hillinger, supra note 212, at 1142-46. 229. 618 P.2d 283, 289 (Haw. 1980), overruled by Francis v. Lee Enters., Inc., 971 P.2d

707, 712-15 (Haw. 1999). 230. Id.

231. Schwartz & Scott, supra note 125, at 545. 232. Id.

233. Id.

456 A DUAL APPROACH TO CONTRACT REMEDIES higher prices to the tiny bookstores if the remedies for breach of its contracts with them expose it to higher damages. Those tiny bookstores, already denied the discounts received by the big chains, 3 4 would be put at an even greater competitive disadvantage. Given the choice, the tiny bookstores probably would prefer to pay a lower price to the publisher and take their chances on being undercompensated in the event of breach. The second problem is transparency. If the cutoff between performance-oriented remedies and option-oriented remedies were based on the number of employees, how would each party learn the size of the other's workforce, and what would happen if the size of the workforce changed during the course of a given contract or series of contracts? Would there be a notice requirement to alert the other party to changes in liability exposure? Any confusion about which system of remedies applies would lead parties to act in reliance on conflicting expectations concerning the applicable background rules."' It would be unfortunate if the dual theory created examples of the party-error problem it was intended to solve. Perhaps, therefore, the simplest and best dividing line would be to provide option-based remedies for all contracts among businesses and to provide performance-based remedies for contracts involving at least one nonbusiness party. 3. Should the System of Remedies Be Contractible?

One solution to the performance/option divide is to let the parties decide which type of remedies should be available for breach. Under this approach, the legislature would set a default-such as option-based remedies for all contracts, performance-based remedies for all contracts, or, as the previous Subsection suggested, option-based remedies for contracts among businesses and performance-based remedies in all other cases. The parties would then be free to bargain around the default to the remedies they prefer. Giving parties the opportunity to choose between contracting for value and contracting for performance would appear to be autonomy-enhancing, but the choice of rules might be only theoretically free. The reason is that a contract not only sets the terms of an agreement but also communicates information about the parties and their expectations. For example, if management suggests adding a profit-sharing term to a labor contract, the union might infer that manage-

234. See, e.g., Am. Booksellers Ass'n, Inc. v. Barnes & Noble, Inc., 135 F. Supp. 2d 1031, 1039 (N.D. Cal. 2001). 235. For an argument that predictability and uniformity can be valuable enough to merit a state's adoption of an inferior rule that has been chosen by the majority of other states, see Northrop Corp. v. Litronic Indus., 29 F.3d 1173, 1178-79 (7 th Cir. 1994) (Posner, J.) (predicting that Illinois would adopt an inferior but majoritari- an interpretation of U.C.C. § 2-207 rather than a better interpretation followed only by California).

457 YALE LAW & POLICY REVIEW 30 :41 5 2012 ment expects a slowdown in earnings." If the promisee seeks the increased protection that contracting for performance would provide, that could be read as a signal of distrust, especially if the promisee is willing to pay for that choice. In another context, Gregory Klass noted that "[ilnterpretive rules that require parties who want, or who do not want, legal liability expressly to say so can in- terfere with and erode extralegal forms of trust that otherwise create value in transactions.""' Still, if erosion of trust is the price for setting clear expectations for performance, the parties' trust might not have been well-founded in the first place. The real problem with contractibility is that too many promisees would agree to option-based contract remedies, whether out of ignorance of the contract terms or out of an optimism bias that causes contractors to discount the likelihood of breach."' An unsophisticated promisee often would be unwilling, at formation, to pay a higher price in exchange for fully compensatory remedies in the event of breach. The promisor usually will prefer a contract with option-oriented remedies, and she will generally be in a position to draft the contract. Making performance-oriented remedies the immutable rule for nonbusiness promisees would protect people who think they are contracting for performance from others who prefer an option rule and have the sophistication to insert such a rule into the contract.239 Immutability does come at a cost, because it prevents the formation of some welfare-enhancing contracts. There will be times when the higher ex ante price that performance-based contract remedies require will be too high for a promisee who, fully understanding the consequences, would have been willing and able to pay the option-contract price. Of course, similar arguments can be made against any required consumer-protection measure. For example, consumers unable to afford a car with air bags might, fully understanding the con-

236. Kathryn E. Spier, Incomplete Contracts and Signalling, 23 RAND J. EcoN. 432, 433 (1992) (providing examples of parties hesitating to request a contractual clause because of the information such a request would appear to convey). 237. Gregory Klass, Intent To Contract, 95 VA. L. REV. 1437, 1441 (2009); see also Omri Ben-Shahar & John A.E. Pottow, On the Stickiness of Default Rules, 33 FLA. ST. U. L. REV. 651, 652-53 (2006) (showing that a party might choose not to ask for a de- viation from the default lest the counterparty impute a negative motive). 238. See supra note 166 and accompanying text. 239. Technically, immutability might not be achievable if the choice of contract theories depends on whether the parties are individuals or businesses, as individuals are free to incorporate. Still, the costs of incorporation would significantly impede unincorporated individuals who wanted to enter contracts with option-oriented remedies. This would create what Ian Ayres refers to as an "impeding altering rule" and would serve the soft-paternalism function of assuring that no individual would accidentally find herself without performance-oriented remedies. See Ayres, supra note 9.

458 A DUAL APPROACH TO CONTRACT REMEDIES sequences, be willing to pay the lower price of a car without air bags. 40 By re- moving the choice of whether to install air bags from the promisor (the car company) and the promisee (the consumer), the government prevents the formation of some Pareto efficient contracts but increases overall auto safety. Making performance-oriented remedies an immutable rule for contracts involving individuals makes sense only if the gains from fully compensating breach outweigh the costs. I believe they would.

CONCLUSION

Empirical evidence raises a problem for contract law: What most people want and expect out of contracts-obligations for performance-differs from what the prevailing theory gives them-options to perform or pay damages. That does not make the option theory "wrong" and the performance theory c right;" both hold descriptive and prescriptive power for different categories of parties. But when what people expect from contract enforcement differs from what they get, the result undermines contract law's legitimacy and efficiency. Legitimacy relies on informed consent, and economic efficiency improves with the accuracy of the parties' understandings about potential outcomes and their likelihoods. The problems of both legitimacy and efficiency would be addressed if parties created liquidated damages clauses and courts enforced them. Neither creation nor enforcement of these clauses is the norm, and neither is likely to become the norm. Trying to educate unsophisticated contractors about the realities of contract law also might be both a losing battle and the wrong one to fight. Better solutions involve recognizing that some parties contract for value and others contract for performance, and that the choice and measure of remedies should account for that difference. For instance, those who contract for performance could receive easier access to specific performance or damages that compensate them not only for the value of what they did not receive, but also for the harm they suffered by receiving money instead of performance. Expecta- tion damages that reflect the subjective value of performance might actually increase economic efficiency by inducing the promisor to factor in all of the harms of breach when deciding whether to perform. Sophisticated parties might not want to absorb the added costs that would come with such contracts, especially when one factors in the litigation expense of determining the reasons for breach and the promisee's valuations and harms. A dual system offers a solution to that issue while preserving the opportunity

240. Not all car owners benefit from air bags, but all new cars come with air bags in- stalled. The government allows people with short legs or certain medical condi- tions to install on/off switches for driver's-side air bags. Deactivation, NHTSA.Gov, http://www.nhtsa.gov/people/injury/airbags/airbagso3/page8.html (last visited Feb. 17, 2012).

459 YALE LAW &POLICY REVIEW 30:415 2012 for unsophisticated parties to be governed by a contract law that respects and delivers on their desires. It is time for the historic dispute between option- and performance-based contract theories to end-not in a verdict for one or the other, but in a settlement that recognizes the value of both.

460